controlnet / lav-df Goto Github PK

View Code? Open in Web Editor NEW

72.0 72.0 9.0 884 KB

[CVIU] Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localization

Home Page: https://www.sciencedirect.com/science/article/pii/S1077314223001984

License: Other

Python 100.00%

cc-by-nc deep-learning deepfakes pytorch

lav-df's Introduction

🎓 I'm a research fellow (postdoc) studying in Computer Vision and Artificial Intelligence area. Now I am mainly working on

Video Analysis
Visual Perceptions
Deepfakes

🔎 Reviewer of IEEE TAFFC, IEEE TMM, ACM TKDD, CVIU, INFFUS, KBS, IEEE TAI, ECCV, ACM MM, ACM ICMI, MBE, IET-CVI, DICTA, CVIP, ICVGIP.

🖥️ I enjoy programming and implementing some cool ideas.

🧰 Also, I love discovering and fine-tuning tools in my hand; both software tools and physical tools (zsh environment, syntax highlighting).

🔔 lllyasviel/ControlNet is a great work and uses the same name, but it's unrelated to me.

💾 Programming Languages and Tools

📮 Contact

lav-df's People

Contributors

Stargazers

Watchers

Forkers

muhammadarslanshahzad i-am-shreya mrizkymunggaran ai-ron-man janfschr asad-be18 myfirstkindom strategist922 airbail

lav-df's Issues

Metadata generation

Hello, I want to test your code on my own videos. Is there a way to generate the required metadata automatically for my own videos? I.e. how did you generate the metadata for LAV-DF? Do you have code for this?

Code related to the deepfake detection task

The dataset you have provided is wonderful for Temporal Forgery Localization. The model proposed for the paper detection is also excellent.
In addition, the paper mentions that it also works well for the basic task of forgery detection, and I have been engaged in research related to multimodal forgery detection for almost half a year, and I don't have much results or ideas now that I'm almost in my third year of research, so it would help me a lot if I can additionally reproduce your work related to forgery detection on top of the original provision of the localization task. Can you provide me with any code or information related to this part of forgery detection? Thank you very much ！！！！
If it's not convenient, just pretend I don't exist.
Have fun.

[email protected]

Frame-level processing

Hello,

In this new Temporal Forgery Localization model, the entire video is used as input. The existing model proposed by the authors has demonstrated promise in achieving accurate results. Although the current input strategy consists of the entire video, it may pose a challenge in terms of memory consumption, especially for large datasets or videos with high resolution frames. Would it be possible to modify the Temporal Forgery Localization model to accept individual frames instead of the entire video? This would result in a reduction in the amount of RAM required.

Run checkpoint on new dataset

Hi,

Thanks for your awesome work. I'm looking to benchmark your checkpoints on a video dataset that I have created. Do you have advice on the easiest way to do this?

Thanks!
Andy

help,where is the metadata.min.json

first,thank you for your wonderful work.
when i run the train.py i met a problem:[Errno 2] No such file or directory: 'E:\LAV-DF\LAV-DF\metadata.min.json'.
i dont understand about renaming the metadata.json to metadata.min.json. because i cant find metadata.json too.
@i-am-shreya @ControlNet
please tell me how to fix it.

Metadata

I hope this message finds you well.
I am currently working on a project and encountered an issue with the Metadata class in the lavdf.py file. When trying to initialize the Metadata class from a JSON file using the lambda function in the read_json method, I received the following error:

TypeError: Metadata.init() got an unexpected keyword argument 'transcript'
Here is the relevant code snippet:
self.metadata: List[Metadata] = read_json(os.path.join(self.root, "metadata.min.json"), lambda x: Metadata(**x))

It seems that the transcript field is not being recognized in the Metadata class, even though I've added it as an attribute. I would greatly appreciate your guidance on resolving this issue.

Thank you very much for your assistance.

Evaluation on my custom dataset

Thank you for your awesome work!
Now, I want to evaluate my own data based on the provided ckpt of batfd_plus. My custom dataset does not have any more information except for the video itself.

Following the code in dataset/lavdf.py, I generate my own metadata.min.json, which is shown as follows:
{ "file": "test/Idexcwshaun1.mp4", "fake_periods": [], "split": "test", "video_frames": 287 }

But I have encountered many errors, such as in dataset/lavdf.py, model/batfd_plus.py.

Or could you help me to inference the single video for deepfake localizations based on batfd_plus?

Looking forward to your reply!
Best wishes.

How to train the MLP classifier on DFDC?

Hi,

In your paper, you said that you trained a MLP classifier using the confidences of predicted segments for deepfake classification on DFDC. I wonder if the MLP classifier is trained separately with confidences, or trained as a backend with the whole audio-visual model?

Liu

Train my own dataset

can i use it to train my own vedio dataset, how can i do

missing meta ?

Hi,

Thanks for your work.

I downloaded the dataset and while I exploring it, I noticed that some video doesn't have a meta entry.

I found 138 010 videos (dev=31925, test=26425, train=79660) but the length of metadata.json is only 136304.

Where can I can find the 1706 missing entries ?

Thank you

["dev/004108.mp4", "dev/004110.mp4", "dev/004120.mp4", "dev/004122.mp4", "dev/004132.mp4", "dev/004134.mp4", "dev/004136.mp4", "dev/004138.mp4", "dev/004148.mp4", "dev/004158.mp4", "dev/004160.mp4", "dev/004166.mp4", "dev/004168.mp4", "dev/004196.mp4", "dev/004199.mp4", "dev/004246.mp4", "dev/004248.mp4", "dev/004315.mp4", "dev/004317.mp4", "dev/004320.mp4", "dev/004339.mp4", "dev/004341.mp4", "dev/004342.mp4", "dev/004344.mp4", "dev/004464.mp4", "dev/004466.mp4", "dev/004467.mp4", "dev/004469.mp4", "dev/004486.mp4", "dev/004488.mp4", "dev/004562.mp4", "dev/004564.mp4", "dev/004575.mp4", "dev/004578.mp4", "dev/004612.mp4", "dev/004614.mp4", "dev/004687.mp4", "dev/004689.mp4", "dev/004699.mp4", "dev/004701.mp4", "dev/004706.mp4", "dev/004708.mp4", "dev/004709.mp4", "dev/004711.mp4", "dev/004732.mp4", "dev/004734.mp4", "dev/004739.mp4", "dev/004741.mp4", "dev/004743.mp4", "dev/004745.mp4", "dev/004758.mp4", "dev/004760.mp4", "dev/004767.mp4", "dev/004769.mp4", "dev/004771.mp4", "dev/004773.mp4", "dev/004775.mp4", "dev/004777.mp4", "dev/004779.mp4", "dev/004781.mp4", "dev/004782.mp4", "dev/004784.mp4", "dev/004786.mp4", "dev/004788.mp4", "dev/004790.mp4", "dev/004792.mp4", "dev/004794.mp4", "dev/004796.mp4", "dev/004809.mp4", "dev/004811.mp4", "dev/004926.mp4", "dev/004928.mp4", "dev/004950.mp4", "dev/004952.mp4", "dev/004953.mp4", "dev/004955.mp4", "dev/004996.mp4", "dev/005032.mp4", "dev/005034.mp4", "dev/005035.mp4", "dev/005037.mp4", "dev/005080.mp4", "dev/005082.mp4", "dev/005083.mp4", "dev/005085.mp4", "dev/005102.mp4", "dev/005104.mp4", "dev/005172.mp4", "dev/005174.mp4", "dev/005319.mp4", "dev/005376.mp4", "dev/005377.mp4", "dev/005378.mp4", "dev/005379.mp4", "dev/005380.mp4", "dev/005381.mp4", "dev/005424.mp4", "dev/005505.mp4", "dev/005507.mp4", "dev/005635.mp4", "dev/005637.mp4", "dev/005638.mp4", "dev/005640.mp4", "dev/005674.mp4", "dev/005676.mp4", "dev/005677.mp4", "dev/005679.mp4", "dev/005692.mp4", "dev/005696.mp4", "dev/005698.mp4", "dev/005703.mp4", "dev/005705.mp4", "dev/005715.mp4", "dev/005717.mp4", "dev/005718.mp4", "dev/005720.mp4", "dev/005722.mp4", "dev/005724.mp4", "dev/005725.mp4", "dev/005727.mp4", "dev/006424.mp4", "dev/006426.mp4", "dev/006452.mp4", "dev/006454.mp4", "dev/011615.mp4", "dev/011618.mp4", "dev/011675.mp4", "dev/011752.mp4", "dev/012364.mp4", "dev/012910.mp4", "dev/016106.mp4", "dev/016108.mp4", "dev/023260.mp4", "dev/023401.mp4", "dev/023717.mp4", "dev/023719.mp4", "dev/023961.mp4", "dev/023963.mp4", "dev/024107.mp4", "dev/024108.mp4", "dev/024109.mp4", "dev/024110.mp4", "dev/024112.mp4", "dev/024328.mp4", "dev/024330.mp4", "dev/024331.mp4", "dev/024333.mp4", "dev/024420.mp4", "dev/024422.mp4", "dev/024585.mp4", "dev/024587.mp4", "dev/024588.mp4", "dev/024590.mp4", "dev/033830.mp4", "dev/033832.mp4", "dev/033849.mp4", "dev/033851.mp4", "dev/034264.mp4", "dev/034266.mp4", "dev/035333.mp4", "dev/035388.mp4", "dev/035506.mp4", "dev/035507.mp4", "dev/035508.mp4", "dev/035509.mp4", "dev/035510.mp4", "dev/035511.mp4", "dev/035512.mp4", "dev/035841.mp4", "dev/035843.mp4", "dev/036173.mp4", "dev/036459.mp4", "dev/036461.mp4", "dev/036953.mp4", "dev/036955.mp4", "dev/036969.mp4", "dev/036970.mp4", "dev/036971.mp4", "dev/036972.mp4", "dev/036973.mp4", "dev/036974.mp4", "dev/037052.mp4", "dev/037053.mp4", "dev/037054.mp4", "dev/037055.mp4", "dev/037056.mp4", "dev/037057.mp4", "dev/037058.mp4", "dev/045410.mp4", "dev/045487.mp4", "dev/045489.mp4", "dev/045596.mp4", "dev/047588.mp4", "dev/050129.mp4", "dev/050282.mp4", "dev/050284.mp4", "dev/054822.mp4", "dev/054824.mp4", "dev/055157.mp4", "dev/055581.mp4", "dev/055583.mp4", "dev/055622.mp4", "dev/055624.mp4", "dev/055817.mp4", "dev/055819.mp4", "dev/057617.mp4", "dev/057903.mp4", "dev/057905.mp4", "dev/058038.mp4", "dev/058040.mp4", "dev/058041.mp4", "dev/058043.mp4", "dev/058372.mp4", "dev/058803.mp4", "dev/058946.mp4", "dev/058948.mp4", "dev/059049.mp4", "dev/059051.mp4", "dev/059200.mp4", "dev/059202.mp4", "dev/059676.mp4", "dev/059678.mp4", "dev/059773.mp4", "dev/059776.mp4", "dev/059983.mp4", "dev/059985.mp4", "dev/059986.mp4", "dev/059988.mp4", "dev/060036.mp4", "dev/060128.mp4", "dev/060304.mp4", "dev/060306.mp4", "dev/060413.mp4", "dev/060415.mp4", "dev/060424.mp4", "dev/060426.mp4", "dev/060427.mp4", "dev/060429.mp4", "dev/060685.mp4", "dev/060687.mp4", "dev/062291.mp4", "dev/062292.mp4", "dev/062293.mp4", "dev/062294.mp4", "dev/062295.mp4", "dev/062296.mp4", "dev/062324.mp4", "dev/062326.mp4", "dev/062327.mp4", "dev/062329.mp4", "dev/062347.mp4", "dev/062349.mp4", "dev/062350.mp4", "dev/062352.mp4", "dev/062584.mp4", "dev/062586.mp4", "dev/062587.mp4", "dev/062589.mp4", "dev/062660.mp4", "dev/062663.mp4", "dev/062665.mp4", "dev/062689.mp4", "dev/062791.mp4", "dev/062793.mp4", "dev/062794.mp4", "dev/062796.mp4", "dev/062855.mp4", "dev/062858.mp4", "dev/062891.mp4", "dev/062893.mp4", "dev/062894.mp4", "dev/062896.mp4", "dev/063003.mp4", "dev/063005.mp4", "dev/063151.mp4", "dev/063152.mp4", "dev/063153.mp4", "dev/063154.mp4", "dev/063155.mp4", "dev/063156.mp4", "dev/063157.mp4", "dev/063467.mp4", "dev/063469.mp4", "dev/063470.mp4", "dev/063472.mp4", "dev/067663.mp4", "dev/067665.mp4", "dev/067666.mp4", "dev/067668.mp4", "dev/068165.mp4", "dev/068167.mp4", "dev/068174.mp4", "dev/068176.mp4", "dev/068454.mp4", "dev/068475.mp4", "dev/068476.mp4", "dev/068477.mp4", "dev/068478.mp4", "dev/068479.mp4", "dev/068480.mp4", "dev/068481.mp4", "dev/070234.mp4", "dev/070258.mp4", "dev/070260.mp4", "dev/070350.mp4", "dev/070353.mp4", "dev/070692.mp4", "dev/071002.mp4", "dev/071004.mp4", "dev/071005.mp4", "dev/071007.mp4", "dev/071061.mp4", "dev/071402.mp4", "dev/071537.mp4", "dev/071620.mp4", "dev/075702.mp4", "dev/075703.mp4", "dev/075704.mp4", "dev/075705.mp4", "dev/075707.mp4", "dev/075728.mp4", "dev/075730.mp4", "dev/075731.mp4", "dev/075733.mp4", "dev/096771.mp4", "dev/097038.mp4", "dev/097414.mp4", "dev/097606.mp4", "dev/097608.mp4", "dev/097609.mp4", "dev/097611.mp4", "dev/099961.mp4", "dev/099963.mp4", "dev/100314.mp4", "dev/100889.mp4", "dev/100891.mp4", "dev/108850.mp4", "dev/108924.mp4", "dev/108935.mp4", "dev/108937.mp4", "dev/109800.mp4", "dev/109802.mp4", "dev/109803.mp4", "dev/109805.mp4", "dev/115422.mp4", "dev/115424.mp4", "dev/116136.mp4", "dev/116138.mp4", "dev/116139.mp4", "dev/116141.mp4", "dev/116301.mp4", "dev/116303.mp4", "dev/116562.mp4", "dev/116564.mp4", "dev/116565.mp4", "dev/116814.mp4", "dev/116817.mp4", "dev/117057.mp4", "dev/117076.mp4", "dev/117089.mp4", "dev/117091.mp4", "dev/117358.mp4", "dev/117360.mp4", "dev/117361.mp4", "dev/117363.mp4", "dev/117374.mp4", "dev/117414.mp4", "dev/117416.mp4", "dev/117451.mp4", "dev/117454.mp4", "dev/117495.mp4", "dev/117497.mp4", "dev/117498.mp4", "dev/117500.mp4", "dev/117650.mp4", "dev/117652.mp4", "dev/117747.mp4", "dev/117749.mp4", "dev/117750.mp4", "dev/117752.mp4", "dev/121298.mp4", "dev/121300.mp4", "dev/121900.mp4", "dev/121902.mp4", "dev/121941.mp4", "dev/121943.mp4", "dev/121972.mp4", "dev/121974.mp4", "dev/121975.mp4", "dev/121977.mp4", "dev/122050.mp4", "dev/122051.mp4", "dev/122052.mp4", "dev/122053.mp4", "dev/122054.mp4", "dev/122055.mp4", "dev/122056.mp4", "dev/122082.mp4", "dev/122084.mp4", "dev/122085.mp4", "dev/123357.mp4", "dev/123359.mp4", "dev/123426.mp4", "dev/123428.mp4", "dev/123501.mp4", "dev/123503.mp4", "dev/123567.mp4", "dev/123569.mp4", "dev/123663.mp4", "dev/123665.mp4", "dev/124168.mp4", "dev/124170.mp4", "dev/132874.mp4", "dev/132924.mp4", "dev/132926.mp4", "dev/133064.mp4", "dev/133256.mp4", "dev/135808.mp4", "dev/135810.mp4", "dev/135811.mp4", "dev/135813.mp4", "dev/135831.mp4", "dev/135833.mp4", "dev/136264.mp4", "test/000007.mp4", "test/000210.mp4", "test/001417.mp4", "test/001484.mp4", "test/001569.mp4", "test/001581.mp4", "test/001583.mp4", "test/001627.mp4", "test/001629.mp4", "test/001630.mp4", "test/001632.mp4", "test/002451.mp4", "test/002453.mp4", "test/002454.mp4", "test/002456.mp4", "test/002483.mp4", "test/002511.mp4", "test/002513.mp4", "test/002669.mp4", "test/002670.mp4", "test/002671.mp4", "test/002745.mp4", "test/002747.mp4", "test/002806.mp4", "test/002808.mp4", "test/002809.mp4", "test/002830.mp4", "test/002832.mp4", "test/003084.mp4", "test/003085.mp4", "test/003086.mp4", "test/003087.mp4", "test/003088.mp4", "test/003089.mp4", "test/003166.mp4", "test/003167.mp4", "test/003168.mp4", "test/003169.mp4", "test/003171.mp4", "test/003265.mp4", "test/003650.mp4", "test/003774.mp4", "test/003776.mp4", "test/003943.mp4", "test/003945.mp4", "test/003946.mp4", "test/003948.mp4", "test/004018.mp4", "test/004020.mp4", "test/005760.mp4", "test/007372.mp4", "test/007374.mp4", "test/007388.mp4", "test/007390.mp4", "test/007461.mp4", "test/007463.mp4", "test/007464.mp4", "test/007466.mp4", "test/007492.mp4", "test/007493.mp4", "test/007494.mp4", "test/007495.mp4", "test/007497.mp4", "test/009528.mp4", "test/009531.mp4", "test/009820.mp4", "test/012472.mp4", "test/012474.mp4", "test/012735.mp4", "test/012745.mp4", "test/012747.mp4", "test/012749.mp4", "test/012781.mp4", "test/012783.mp4", "test/012784.mp4", "test/012786.mp4", "test/012821.mp4", "test/022296.mp4", "test/022580.mp4", "test/022582.mp4", "test/022876.mp4", "test/022878.mp4", "test/026541.mp4", "test/026543.mp4", "test/026548.mp4", "test/026549.mp4", "test/026550.mp4", "test/026551.mp4", "test/026552.mp4", "test/026553.mp4", "test/026759.mp4", "test/026761.mp4", "test/026941.mp4", "test/027108.mp4", "test/027187.mp4", "test/027189.mp4", "test/027190.mp4", "test/027192.mp4", "test/027194.mp4", "test/027196.mp4", "test/027197.mp4", "test/027199.mp4", "test/027245.mp4", "test/027247.mp4", "test/027252.mp4", "test/027254.mp4", "test/029355.mp4", "test/029357.mp4", "test/041196.mp4", "test/041198.mp4", "test/044990.mp4", "test/045118.mp4", "test/045120.mp4", "test/045121.mp4", "test/045123.mp4", "test/045162.mp4", "test/045164.mp4", "test/045775.mp4", "test/045965.mp4", "test/045967.mp4", "test/046074.mp4", "test/046327.mp4", "test/046329.mp4", "test/048182.mp4", "test/048248.mp4", "test/048250.mp4", "test/063607.mp4", "test/063609.mp4", "test/063952.mp4", "test/063955.mp4", "test/064114.mp4", "test/064155.mp4", "test/075955.mp4", "test/076013.mp4", "test/076015.mp4", "test/076018.mp4", "test/076412.mp4", "test/076413.mp4", "test/076414.mp4", "test/076415.mp4", "test/076416.mp4", "test/076417.mp4", "test/076418.mp4", "test/076540.mp4", "test/076542.mp4", "test/076543.mp4", "test/076545.mp4", "test/076695.mp4", "test/076697.mp4", "test/076698.mp4", "test/076700.mp4", "test/076702.mp4", "test/076704.mp4", "test/076854.mp4", "test/076855.mp4", "test/076856.mp4", "test/076857.mp4", "test/076859.mp4", "test/076893.mp4", "test/076894.mp4", "test/076895.mp4", "test/076896.mp4", "test/076897.mp4", "test/076898.mp4", "test/076899.mp4", "test/076925.mp4", "test/077338.mp4", "test/077340.mp4", "test/077497.mp4", "test/077499.mp4", "test/077500.mp4", "test/077502.mp4", "test/077716.mp4", "test/078456.mp4", "test/090135.mp4", "test/090138.mp4", "test/090140.mp4", "test/090260.mp4", "test/090262.mp4", "test/090263.mp4", "test/090265.mp4", "test/090268.mp4", "test/090269.mp4", "test/090270.mp4", "test/090271.mp4", "test/090273.mp4", "test/090838.mp4", "test/090840.mp4", "test/090872.mp4", "test/090874.mp4", "test/091055.mp4", "test/091057.mp4", "test/091131.mp4", "test/091149.mp4", "test/091150.mp4", "test/091151.mp4", "test/091152.mp4", "test/091154.mp4", "test/091248.mp4", "test/091483.mp4", "test/091485.mp4", "test/091769.mp4", "test/091771.mp4", "test/094695.mp4", "test/094697.mp4", "test/095029.mp4", "test/095031.mp4", "test/095032.mp4", "test/095034.mp4", "test/095049.mp4", "test/095431.mp4", "test/095432.mp4", "test/095433.mp4", "test/095434.mp4", "test/095435.mp4", "test/095436.mp4", "test/095486.mp4", "test/095488.mp4", "test/095489.mp4", "test/095491.mp4", "test/095616.mp4", "test/095618.mp4", "test/095962.mp4", "test/095964.mp4", "test/096208.mp4", "test/104957.mp4", "test/104964.mp4", "test/104966.mp4", "test/105587.mp4", "test/105589.mp4", "test/105643.mp4", "test/105645.mp4", "test/105663.mp4", "test/105665.mp4", "test/114123.mp4", "test/114252.mp4", "test/114254.mp4", "test/114704.mp4", "test/125390.mp4", "test/125392.mp4", "test/125817.mp4", "test/125819.mp4", "test/125820.mp4", "test/125832.mp4", "test/125834.mp4", "test/125957.mp4", "test/125959.mp4", "test/126268.mp4", "test/126304.mp4", "test/126306.mp4", "test/126502.mp4", "test/126503.mp4", "test/126504.mp4", "test/126505.mp4", "test/126507.mp4", "test/129637.mp4", "test/129640.mp4", "test/129932.mp4", "test/129935.mp4", "test/130230.mp4", "test/130246.mp4", "test/130248.mp4", "test/130407.mp4", "test/130409.mp4", "test/130410.mp4", "test/130412.mp4", "test/130555.mp4", "test/131001.mp4", "test/131057.mp4", "test/131059.mp4", "test/131232.mp4", "test/131351.mp4", "test/131585.mp4", "test/131587.mp4", "test/131760.mp4", "test/131761.mp4", "test/131762.mp4", "test/131763.mp4", "test/131765.mp4", "test/131771.mp4", "test/131773.mp4", "test/131788.mp4", "test/131789.mp4", "test/131790.mp4", "test/131791.mp4", "test/131792.mp4", "test/131793.mp4", "test/131794.mp4", "test/131811.mp4", "test/131876.mp4", "test/131879.mp4", "test/131967.mp4", "test/131968.mp4", "test/131969.mp4", "test/131970.mp4", "test/131971.mp4", "test/131972.mp4", "test/132030.mp4", "test/132032.mp4", "test/132033.mp4", "test/132035.mp4", "test/133946.mp4", "test/133948.mp4", "test/133996.mp4", "test/133998.mp4", "test/134169.mp4", "test/134170.mp4", "test/134171.mp4", "test/134172.mp4", "test/134173.mp4", "test/134174.mp4", "test/134175.mp4", "test/134258.mp4", "test/134260.mp4", "test/134262.mp4", "test/134264.mp4", "test/134288.mp4", "test/135056.mp4", "test/135075.mp4", "test/135164.mp4", "test/135166.mp4", "test/135168.mp4", "test/135170.mp4", "test/135184.mp4", "test/135186.mp4", "train/000866.mp4", "train/000868.mp4", "train/006287.mp4", "train/008087.mp4", "train/008089.mp4", "train/008473.mp4", "train/008475.mp4", "train/008718.mp4", "train/008719.mp4", "train/008720.mp4", "train/008721.mp4", "train/008722.mp4", "train/008723.mp4", "train/008730.mp4", "train/008732.mp4", "train/008994.mp4", "train/008995.mp4", "train/008996.mp4", "train/008997.mp4", "train/008998.mp4", "train/008999.mp4", "train/009248.mp4", "train/009250.mp4", "train/009389.mp4", "train/010231.mp4", "train/011185.mp4", "train/011199.mp4", "train/011201.mp4", "train/011203.mp4", "train/011205.mp4", "train/011207.mp4", "train/011209.mp4", "train/011222.mp4", "train/011224.mp4", "train/011320.mp4", "train/011322.mp4", "train/011323.mp4", "train/011325.mp4", "train/011433.mp4", "train/011435.mp4", "train/011436.mp4", "train/011438.mp4", "train/013217.mp4", "train/013627.mp4", "train/013629.mp4", "train/013630.mp4", "train/013632.mp4", "train/014015.mp4", "train/014018.mp4", "train/014189.mp4", "train/014190.mp4", "train/014191.mp4", "train/014192.mp4", "train/014194.mp4", "train/014265.mp4", "train/014267.mp4", "train/014307.mp4", "train/014309.mp4", "train/014310.mp4", "train/014312.mp4", "train/014356.mp4", "train/014358.mp4", "train/014466.mp4", "train/014661.mp4", "train/014663.mp4", "train/014664.mp4", "train/014666.mp4", "train/014926.mp4", "train/014928.mp4", "train/015115.mp4", "train/015168.mp4", "train/015170.mp4", "train/015171.mp4", "train/015173.mp4", "train/015183.mp4", "train/015185.mp4", "train/015204.mp4", "train/015206.mp4", "train/015298.mp4", "train/015374.mp4", "train/015376.mp4", "train/015448.mp4", "train/015460.mp4", "train/015462.mp4", "train/015464.mp4", "train/015466.mp4", "train/015467.mp4", "train/015469.mp4", "train/015471.mp4", "train/015473.mp4", "train/015474.mp4", "train/015476.mp4", "train/015478.mp4", "train/015489.mp4", "train/015491.mp4", "train/015493.mp4", "train/015495.mp4", "train/015497.mp4", "train/015592.mp4", "train/015594.mp4", "train/015596.mp4", "train/015598.mp4", "train/015646.mp4", "train/015647.mp4", "train/015648.mp4", "train/015649.mp4", "train/015650.mp4", "train/015651.mp4", "train/016742.mp4", "train/016743.mp4", "train/016744.mp4", "train/016745.mp4", "train/016747.mp4", "train/016779.mp4", "train/016781.mp4", "train/016943.mp4", "train/016945.mp4", "train/016992.mp4", "train/017030.mp4", "train/017032.mp4", "train/017310.mp4", "train/017312.mp4", "train/017313.mp4", "train/017315.mp4", "train/017404.mp4", "train/017406.mp4", "train/017458.mp4", "train/017554.mp4", "train/017556.mp4", "train/018058.mp4", "train/018060.mp4", "train/018061.mp4", "train/018063.mp4", "train/018076.mp4", "train/018078.mp4", "train/018200.mp4", "train/018202.mp4", "train/018203.mp4", "train/018374.mp4", "train/018377.mp4", "train/018391.mp4", "train/018393.mp4", "train/018395.mp4", "train/018397.mp4", "train/018400.mp4", "train/018402.mp4", "train/018404.mp4", "train/018406.mp4", "train/018414.mp4", "train/018416.mp4", "train/018421.mp4", "train/018422.mp4", "train/018423.mp4", "train/018440.mp4", "train/018450.mp4", "train/018472.mp4", "train/018474.mp4", "train/018484.mp4", "train/018486.mp4", "train/018489.mp4", "train/018491.mp4", "train/018492.mp4", "train/018514.mp4", "train/018516.mp4", "train/018518.mp4", "train/018520.mp4", "train/018523.mp4", "train/018525.mp4", "train/018526.mp4", "train/018528.mp4", "train/018531.mp4", "train/018533.mp4", "train/018544.mp4", "train/018546.mp4", "train/018548.mp4", "train/018550.mp4", "train/018552.mp4", "train/018554.mp4", "train/018558.mp4", "train/018562.mp4", "train/018564.mp4", "train/018566.mp4", "train/018568.mp4", "train/018570.mp4", "train/018572.mp4", "train/018574.mp4", "train/018576.mp4", "train/018577.mp4", "train/018579.mp4", "train/018586.mp4", "train/018588.mp4", "train/018590.mp4", "train/018592.mp4", "train/018938.mp4", "train/019049.mp4", "train/019140.mp4", "train/019142.mp4", "train/019144.mp4", "train/019146.mp4", "train/019404.mp4", "train/019407.mp4", "train/019579.mp4", "train/019581.mp4", "train/020194.mp4", "train/020403.mp4", "train/020406.mp4", "train/020563.mp4", "train/020565.mp4", "train/020630.mp4", "train/020632.mp4", "train/020633.mp4", "train/020635.mp4", "train/020859.mp4", "train/020861.mp4", "train/020947.mp4", "train/020948.mp4", "train/020949.mp4", "train/020950.mp4", "train/020951.mp4", "train/020952.mp4", "train/020953.mp4", "train/021255.mp4", "train/021297.mp4", "train/021332.mp4", "train/021334.mp4", "train/021484.mp4", "train/021486.mp4", "train/021734.mp4", "train/021736.mp4", "train/021754.mp4", "train/021756.mp4", "train/021789.mp4", "train/021791.mp4", "train/024915.mp4", "train/025330.mp4", "train/025438.mp4", "train/025440.mp4", "train/025638.mp4", "train/025640.mp4", "train/025675.mp4", "train/025677.mp4", "train/026058.mp4", "train/026060.mp4", "train/026061.mp4", "train/026239.mp4", "train/026241.mp4", "train/026242.mp4", "train/026244.mp4", "train/026374.mp4", "train/026438.mp4", "train/026440.mp4", "train/026441.mp4", "train/026443.mp4", "train/027313.mp4", "train/027314.mp4", "train/027315.mp4", "train/027928.mp4", "train/027930.mp4", "train/028654.mp4", "train/028656.mp4", "train/028657.mp4", "train/028659.mp4", "train/029085.mp4", "train/029088.mp4", "train/029128.mp4", "train/030110.mp4", "train/030112.mp4", "train/030129.mp4", "train/030132.mp4", "train/030317.mp4", "train/030319.mp4", "train/030335.mp4", "train/030441.mp4", "train/030695.mp4", "train/030697.mp4", "train/030720.mp4", "train/030882.mp4", "train/030988.mp4", "train/031170.mp4", "train/031187.mp4", "train/031203.mp4", "train/031329.mp4", "train/031513.mp4", "train/031515.mp4", "train/032693.mp4", "train/032766.mp4", "train/032768.mp4", "train/032785.mp4", "train/033337.mp4", "train/033531.mp4", "train/033585.mp4", "train/033587.mp4", "train/033625.mp4", "train/033627.mp4", "train/033671.mp4", "train/033674.mp4", "train/034390.mp4", "train/034439.mp4", "train/034686.mp4", "train/034687.mp4", "train/034688.mp4", "train/034689.mp4", "train/034690.mp4", "train/034691.mp4", "train/034800.mp4", "train/034893.mp4", "train/034998.mp4", "train/035000.mp4", "train/035143.mp4", "train/035145.mp4", "train/037119.mp4", "train/037133.mp4", "train/037135.mp4", "train/037485.mp4", "train/037486.mp4", "train/037487.mp4", "train/037488.mp4", "train/037489.mp4", "train/037490.mp4", "train/037491.mp4", "train/038100.mp4", "train/038101.mp4", "train/038102.mp4", "train/038103.mp4", "train/038243.mp4", "train/038245.mp4", "train/038246.mp4", "train/038248.mp4", "train/038334.mp4", "train/038349.mp4", "train/038351.mp4", "train/039332.mp4", "train/039559.mp4", "train/039663.mp4", "train/039665.mp4", "train/039666.mp4", "train/039668.mp4", "train/039835.mp4", "train/039837.mp4", "train/039846.mp4", "train/039848.mp4", "train/039947.mp4", "train/039948.mp4", "train/039949.mp4", "train/039950.mp4", "train/039952.mp4", "train/040242.mp4", "train/040290.mp4", "train/040345.mp4", "train/040347.mp4", "train/040531.mp4", "train/040532.mp4", "train/040533.mp4", "train/040534.mp4", "train/040535.mp4", "train/040536.mp4", "train/040542.mp4", "train/040544.mp4", "train/040545.mp4", "train/040547.mp4", "train/040569.mp4", "train/040571.mp4", "train/040572.mp4", "train/040574.mp4", "train/040632.mp4", "train/040633.mp4", "train/040634.mp4", "train/040635.mp4", "train/040637.mp4", "train/040822.mp4", "train/040824.mp4", "train/040946.mp4", "train/041625.mp4", "train/041666.mp4", "train/041700.mp4", "train/042177.mp4", "train/042473.mp4", "train/042535.mp4", "train/042669.mp4", "train/042746.mp4", "train/042828.mp4", "train/042830.mp4", "train/042977.mp4", "train/044344.mp4", "train/048843.mp4", "train/048991.mp4", "train/049396.mp4", "train/049397.mp4", "train/049398.mp4", "train/049399.mp4", "train/049400.mp4", "train/049401.mp4", "train/049586.mp4", "train/049588.mp4", "train/050777.mp4", "train/050778.mp4", "train/050779.mp4", "train/050780.mp4", "train/050782.mp4", "train/051549.mp4", "train/052697.mp4", "train/052699.mp4", "train/052709.mp4", "train/052711.mp4", "train/052716.mp4", "train/052718.mp4", "train/052719.mp4", "train/052721.mp4", "train/052979.mp4", "train/052981.mp4", "train/052982.mp4", "train/052984.mp4", "train/053004.mp4", "train/053005.mp4", "train/053006.mp4", "train/053007.mp4", "train/053009.mp4", "train/053011.mp4", "train/053013.mp4", "train/053014.mp4", "train/053016.mp4", "train/053018.mp4", "train/053020.mp4", "train/053021.mp4", "train/053023.mp4", "train/053025.mp4", "train/053027.mp4", "train/053118.mp4", "train/053147.mp4", "train/053149.mp4", "train/053169.mp4", "train/053201.mp4", "train/053203.mp4", "train/053204.mp4", "train/053206.mp4", "train/053351.mp4", "train/053353.mp4", "train/053366.mp4", "train/053368.mp4", "train/053369.mp4", "train/053371.mp4", "train/053467.mp4", "train/053546.mp4", "train/053548.mp4", "train/053549.mp4", "train/053551.mp4", "train/053653.mp4", "train/053655.mp4", "train/053678.mp4", "train/053680.mp4", "train/053763.mp4", "train/054077.mp4", "train/054148.mp4", "train/056129.mp4", "train/056131.mp4", "train/056308.mp4", "train/056310.mp4", "train/056715.mp4", "train/056981.mp4", "train/056983.mp4", "train/056984.mp4", "train/057114.mp4", "train/057115.mp4", "train/057116.mp4", "train/057117.mp4", "train/057118.mp4", "train/057119.mp4", "train/057159.mp4", "train/057161.mp4", "train/061008.mp4", "train/061011.mp4", "train/061056.mp4", "train/061174.mp4", "train/061176.mp4", "train/061522.mp4", "train/061524.mp4", "train/061525.mp4", "train/061527.mp4", "train/061553.mp4", "train/061960.mp4", "train/062087.mp4", "train/062089.mp4", "train/062125.mp4", "train/064323.mp4", "train/064325.mp4", "train/064433.mp4", "train/064436.mp4", "train/064461.mp4", "train/064463.mp4", "train/064525.mp4", "train/064528.mp4", "train/064591.mp4", "train/064592.mp4", "train/064593.mp4", "train/064594.mp4", "train/064595.mp4", "train/064596.mp4", "train/064597.mp4", "train/064791.mp4", "train/065060.mp4", "train/065069.mp4", "train/065071.mp4", "train/065128.mp4", "train/065181.mp4", "train/065183.mp4", "train/065199.mp4", "train/065277.mp4", "train/065293.mp4", "train/065364.mp4", "train/065366.mp4", "train/065383.mp4", "train/065385.mp4", "train/065468.mp4", "train/065470.mp4", "train/065471.mp4", "train/065473.mp4", "train/065510.mp4", "train/065535.mp4", "train/065537.mp4", "train/065701.mp4", "train/065963.mp4", "train/066268.mp4", "train/066271.mp4", "train/066424.mp4", "train/066500.mp4", "train/066503.mp4", "train/066505.mp4", "train/066506.mp4", "train/066507.mp4", "train/066508.mp4", "train/066592.mp4", "train/066633.mp4", "train/066796.mp4", "train/066814.mp4", "train/066952.mp4", "train/067075.mp4", "train/067078.mp4", "train/069063.mp4", "train/069064.mp4", "train/069065.mp4", "train/069066.mp4", "train/069067.mp4", "train/069068.mp4", "train/069092.mp4", "train/069094.mp4", "train/069315.mp4", "train/069317.mp4", "train/069318.mp4", "train/069320.mp4", "train/069997.mp4", "train/069998.mp4", "train/069999.mp4", "train/070000.mp4", "train/070001.mp4", "train/070002.mp4", "train/070066.mp4", "train/070068.mp4", "train/070089.mp4", "train/070091.mp4", "train/070155.mp4", "train/070156.mp4", "train/070157.mp4", "train/070158.mp4", "train/070160.mp4", "train/072198.mp4", "train/072199.mp4", "train/072200.mp4", "train/072201.mp4", "train/072203.mp4", "train/072267.mp4", "train/072269.mp4", "train/072335.mp4", "train/072425.mp4", "train/072454.mp4", "train/072456.mp4", "train/072457.mp4", "train/072873.mp4", "train/073199.mp4", "train/073200.mp4", "train/073201.mp4", "train/073202.mp4", "train/073203.mp4", "train/073204.mp4", "train/073258.mp4", "train/073260.mp4", "train/073335.mp4", "train/073337.mp4", "train/073514.mp4", "train/075046.mp4", "train/075048.mp4", "train/079208.mp4", "train/079210.mp4", "train/079412.mp4", "train/079413.mp4", "train/079414.mp4", "train/079415.mp4", "train/079416.mp4", "train/079417.mp4", "train/079418.mp4", "train/079485.mp4", "train/079644.mp4", "train/079645.mp4", "train/079646.mp4", "train/079647.mp4", "train/079648.mp4", "train/079649.mp4", "train/080035.mp4", "train/080204.mp4", "train/080206.mp4", "train/080606.mp4", "train/080608.mp4", "train/080609.mp4", "train/080611.mp4", "train/080717.mp4", "train/080719.mp4", "train/080799.mp4", "train/080961.mp4", "train/080963.mp4", "train/081025.mp4", "train/081204.mp4", "train/081788.mp4", "train/081790.mp4", "train/081791.mp4", "train/081793.mp4", "train/081879.mp4", "train/081881.mp4", "train/082290.mp4", "train/082292.mp4", "train/082346.mp4", "train/082692.mp4", "train/082694.mp4", "train/082695.mp4", "train/082697.mp4", "train/083113.mp4", "train/083195.mp4", "train/083197.mp4", "train/083198.mp4", "train/083200.mp4", "train/083524.mp4", "train/083526.mp4", "train/083531.mp4", "train/083533.mp4", "train/083534.mp4", "train/083536.mp4", "train/084209.mp4", "train/084211.mp4", "train/085076.mp4", "train/085211.mp4", "train/085213.mp4", "train/085310.mp4", "train/085381.mp4", "train/085580.mp4", "train/085582.mp4", "train/085701.mp4", "train/085703.mp4", "train/085925.mp4", "train/085928.mp4", "train/085965.mp4", "train/085987.mp4", "train/085989.mp4", "train/085992.mp4", "train/086211.mp4", "train/086218.mp4", "train/086220.mp4", "train/086399.mp4", "train/086477.mp4", "train/086479.mp4", "train/086536.mp4", "train/086538.mp4", "train/086895.mp4", "train/086897.mp4", "train/087286.mp4", "train/087639.mp4", "train/087640.mp4", "train/087641.mp4", "train/087642.mp4", "train/087643.mp4", "train/087644.mp4", "train/087645.mp4", "train/087789.mp4", "train/087813.mp4", "train/087873.mp4", "train/087875.mp4", "train/088403.mp4", "train/088405.mp4", "train/088479.mp4", "train/088481.mp4", "train/088828.mp4", "train/089247.mp4", "train/089388.mp4", "train/089824.mp4", "train/089826.mp4", "train/089923.mp4", "train/089925.mp4", "train/089926.mp4", "train/089928.mp4", "train/090021.mp4", "train/090023.mp4", "train/090024.mp4", "train/090026.mp4", "train/092132.mp4", "train/092391.mp4", "train/092508.mp4", "train/092595.mp4", "train/092924.mp4", "train/092926.mp4", "train/092958.mp4", "train/092991.mp4", "train/093112.mp4", "train/093115.mp4", "train/093623.mp4", "train/093742.mp4", "train/093745.mp4", "train/093825.mp4", "train/093826.mp4", "train/093827.mp4", "train/093828.mp4", "train/093829.mp4", "train/093830.mp4", "train/094003.mp4", "train/094006.mp4", "train/094172.mp4", "train/094174.mp4", "train/094180.mp4", "train/094182.mp4", "train/094185.mp4", "train/094450.mp4", "train/094453.mp4", "train/098215.mp4", "train/098495.mp4", "train/098508.mp4", "train/098511.mp4", "train/098575.mp4", "train/098633.mp4", "train/098646.mp4", "train/098648.mp4", "train/098649.mp4", "train/098651.mp4", "train/098653.mp4", "train/098655.mp4", "train/098833.mp4", "train/099373.mp4", "train/099375.mp4", "train/101212.mp4", "train/101215.mp4", "train/101550.mp4", "train/101552.mp4", "train/101606.mp4", "train/101608.mp4", "train/101724.mp4", "train/101726.mp4", "train/101727.mp4", "train/101729.mp4", "train/101849.mp4", "train/101851.mp4", "train/101852.mp4", "train/101854.mp4", "train/102354.mp4", "train/102355.mp4", "train/102356.mp4", "train/102357.mp4", "train/102359.mp4", "train/103006.mp4", "train/103007.mp4", "train/103008.mp4", "train/103009.mp4", "train/103011.mp4", "train/103090.mp4", "train/103154.mp4", "train/103156.mp4", "train/103163.mp4", "train/103165.mp4", "train/103185.mp4", "train/103230.mp4", "train/103705.mp4", "train/103706.mp4", "train/103707.mp4", "train/103708.mp4", "train/103710.mp4", "train/103910.mp4", "train/103912.mp4", "train/104444.mp4", "train/104543.mp4", "train/104545.mp4", "train/106757.mp4", "train/106759.mp4", "train/107134.mp4", "train/107146.mp4", "train/107147.mp4", "train/107148.mp4", "train/107149.mp4", "train/107151.mp4", "train/107169.mp4", "train/107171.mp4", "train/107172.mp4", "train/107174.mp4", "train/107176.mp4", "train/107178.mp4", "train/107180.mp4", "train/107182.mp4", "train/107183.mp4", "train/107185.mp4", "train/107195.mp4", "train/107197.mp4", "train/107220.mp4", "train/107267.mp4", "train/107442.mp4", "train/107444.mp4", "train/107496.mp4", "train/107599.mp4", "train/107750.mp4", "train/107826.mp4", "train/107827.mp4", "train/107828.mp4", "train/107829.mp4", "train/107831.mp4", "train/108106.mp4", "train/108196.mp4", "train/108499.mp4", "train/108527.mp4", "train/108594.mp4", "train/108596.mp4", "train/109932.mp4", "train/110011.mp4", "train/110014.mp4", "train/110297.mp4", "train/110426.mp4", "train/110569.mp4", "train/111941.mp4", "train/111943.mp4", "train/112146.mp4", "train/112149.mp4", "train/112267.mp4", "train/112269.mp4", "train/112430.mp4", "train/112440.mp4", "train/112459.mp4", "train/112549.mp4", "train/112550.mp4", "train/112551.mp4", "train/112552.mp4", "train/112553.mp4", "train/112554.mp4", "train/112889.mp4", "train/112891.mp4", "train/113024.mp4", "train/113026.mp4", "train/113295.mp4", "train/113297.mp4", "train/113373.mp4", "train/113380.mp4", "train/113382.mp4", "train/113682.mp4", "train/113684.mp4", "train/113685.mp4", "train/113687.mp4", "train/113823.mp4", "train/114498.mp4", "train/117853.mp4", "train/117855.mp4", "train/117858.mp4", "train/117923.mp4", "train/117999.mp4", "train/118094.mp4", "train/118096.mp4", "train/118121.mp4", "train/118123.mp4", "train/118197.mp4", "train/118199.mp4", "train/118268.mp4", "train/118270.mp4", "train/118271.mp4", "train/118376.mp4", "train/118378.mp4", "train/119020.mp4", "train/119034.mp4", "train/119036.mp4", "train/119386.mp4", "train/119388.mp4", "train/119483.mp4", "train/119485.mp4", "train/119498.mp4", "train/119500.mp4", "train/119501.mp4", "train/119503.mp4", "train/119505.mp4", "train/119507.mp4", "train/119508.mp4", "train/119607.mp4", "train/119609.mp4", "train/119823.mp4", "train/119825.mp4", "train/119841.mp4", "train/119842.mp4", "train/119843.mp4", "train/119844.mp4", "train/119845.mp4", "train/119846.mp4", "train/119847.mp4", "train/119892.mp4", "train/119894.mp4", "train/120013.mp4", "train/120015.mp4", "train/120084.mp4", "train/120087.mp4", "train/120347.mp4", "train/120349.mp4", "train/120368.mp4", "train/120370.mp4", "train/120372.mp4", "train/120374.mp4", "train/120515.mp4", "train/120517.mp4", "train/120593.mp4", "train/120595.mp4", "train/120678.mp4", "train/120679.mp4", "train/120680.mp4", "train/120681.mp4", "train/120682.mp4", "train/120683.mp4", "train/122656.mp4", "train/122658.mp4", "train/122737.mp4", "train/122739.mp4", "train/122841.mp4", "train/122842.mp4", "train/122843.mp4", "train/122844.mp4", "train/122846.mp4", "train/124506.mp4", "train/124508.mp4", "train/124545.mp4", "train/124547.mp4", "train/124646.mp4", "train/124647.mp4", "train/124648.mp4", "train/124649.mp4", "train/124650.mp4", "train/124651.mp4", "train/124996.mp4", "train/127109.mp4", "train/127111.mp4", "train/128198.mp4", "train/128200.mp4", "train/128917.mp4", "train/129373.mp4", "train/132137.mp4", "train/132554.mp4", "train/135364.mp4", "train/135520.mp4", "train/135610.mp4", "train/136372.mp4", "train/136807.mp4", "train/137022.mp4", "train/137024.mp4", "train/137960.mp4"]

Facing problem with preprocessing for the single video inference

I need help how to preprocess the video and generate the meta file. So I can use the batfd model for single video inference. and to perform the classification

Where to find metadata in LAV-DF dataset?

Dear Thank you for your amazing work.
I have been trying to train the network with the LAV-DF dataset, however, I cannot find the metadata.json and metadata.min.json files in the directory to train the network. Is there anything we can do to generate these files or is there anywhere I can get those files? Would be appreciated.

Thanks

evaluation error

When I run the following command:
python evaluate.py
--config ./config/batfd_default.toml
--data_root $DATASET_PATH
--checkpoint ./ckpt/batfd_default/last.ckpt

I meet this error:

Load 113541 data in train.
Load 0 data in dev.
Load 4150 data in test.
GPU available: True (cuda), used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
Traceback (most recent call last):
File "/opt/data/private/code/DFD/3.LAV-DF-master/evaluate.py", line 128, in
evaluate_lavdf(config, args)
File "/opt/data/private/code/DFD/3.LAV-DF-master/evaluate.py", line 78, in evaluate_lavdf
inference_batfd(model_name, model, dm, config["max_duration"], model_type, args.modalities, args.gpus)
File "/opt/data/private/code/DFD/3.LAV-DF-master/inference.py", line 147, in inference_batfd
trainer.predict(model, dm.test_dataloader())
File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 949, in predict
return self._call_and_handle_interrupt(
File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 648, in _call_and_handle_interrupt
return self.strategy.launcher.launch(trainer_fn, *args, trainer=self, **kwargs)
File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/pytorch_lightning/strategies/launchers/multiprocessing.py", line 107, in launch
mp.start_processes(
File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/torch/multiprocessing/spawn.py", line 189, in start_processes
process.start()
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/process.py", line 121, in start
self._popen = self._Popen(self)
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/context.py", line 284, in _Popen
return Popen(process_obj)
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 32, in init
super().init(process_obj)
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/popen_fork.py", line 19, in init
self._launch(process_obj)
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 47, in _launch
reduction.dump(process_obj, fp)
File "/root/miniconda3/envs/py3.9/lib/python3.9/multiprocessing/reduction.py", line 60, in dump
ForkingPickler(file, protocol).dump(obj)
AttributeError: Can't pickle local object '_get_Conv.._ConvNd'

Could you help me solve this error?

Find a wrong

Thank you for your excellent work！I see you made some changes to the file yesterday，but there's a mistake with that. In the file of “utils”，in 231 lines，you should add “ del meta[“transcript”] , metadata_min.append(meta) ”，then you can get a right metadata.min.json. Otherwise, the metadata.min.json will be null.

Testing LAV-DF metadata.min.json

Hello , I have several forged face video. And I would like to test them on LAVDF.

Can i run this without audio file ?

And When I give video path , it gives "LAVDF\LAV-DF\utils.py", line 16, in read_json
with open(path, 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: './db/metadata.min.json' " error.

What is "metadata.min.json", how to create one?

Thanks for your work.

dataset structure

Hi, its an awesome project. But I have faced some issues. How the dataset structure should be placed in the script? bcz when I download the dataset it's going to be like.zip

检测单个视频

请问如何使用模型去定位单个视频的伪造片段

GPU Requirements for Training

I am attempting to train BA-TFD+ on a custom data set, but I am running into memory issues. I am using a VM with 2 M60 GPUs with 8 GB of memory each, but I am not able to run the training module without crashing, even when setting the batch size to 1. Is there a recommended amount of GPU memory to be able to train the model?

Pretrained Models

Hello,

I would like to thank you for your excellent work. However, the link that you provided for accessing the pretrained models seems to be inactive.

classification results

Congratulations on your excellent work!

I didn't find the code for the classification results and AUC calculation when I re-implemented your code.

Could you share this part of the code?

How many epochs did you train?

Hi,

Your work is very interesting. I am rebuliding your model and I found the training is time-consuming for me using one 3090 GPU. So I want to know the number of gpus you used and the actual epochs when to stop training.

Thank you.

evaluation error

File "/home/pradeep/.local/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 401, in init
self._accelerator_connector = _AcceleratorConnector(
File "/home/pradeep/.local/lib/python3.10/site-packages/pytorch_lightning/trainer/connectors/accelerator_connector.py", line 155, in init
self._set_parallel_devices_and_init_accelerator()
File "/home/pradeep/.local/lib/python3.10/site-packages/pytorch_lightning/trainer/connectors/accelerator_connector.py", line 395, in _set_parallel_devices_and_init_accelerator
self._devices_flag = accelerator_cls.parse_devices(self._devices_flag)
File "/home/pradeep/.local/lib/python3.10/site-packages/pytorch_lightning/accelerators/cuda.py", line 82, in parse_devices
return _parse_gpu_ids(devices, include_cuda=True)
File "/home/pradeep/.local/lib/python3.10/site-packages/lightning_fabric/utilities/device_parser.py", line 77, in _parse_gpu_ids
_check_data_type(gpus)
File "/home/pradeep/.local/lib/python3.10/site-packages/lightning_fabric/utilities/device_parser.py", line 198, in _check_data_type
raise TypeError(f"{msg} None")
TypeError: Device IDs (GPU/TPU) must be an int, a string, a sequence of ints, but you passed None

can any explain me how to solve this error

Unable to change maximum segment duration for BA-TFD

Hello, I am trying to change the max segment duration for the BA-TFD model (not plus) and I running into issues when doing so. I have changed the value in the config file, and everywhere in the codebase where the max_duration variable is manually set to 40, However, I am still getting the following error:

RuntimeError: The size of tensor a (100) must match the size of tensor b (40) at non-singleton dimension 0

Are there additional files that must be edited to change this value?

visualize the result

I am able to run the evaluate.py file with given checkpoint value.After this it created a output folder which contains the result folder
with .csv files. can you please tell me how to use this files to get the output in Videos

Question about boundary sample map weight

Hi, thanks for the great work, I have a question about the function get_pem_smp_weight() of model.boundary_module.py.

At line 77 you assigned xmax = (j + 1), and at line 79 length = xmax - xmin. Based on your paper and the comment written in the file, I thought that xmax should be (i + j + 1) and length = (j + 1).

Consider a case that i = 10, and j = 0, in this case, xmax = 1, length = -9. it is quite confusing to me that how length < 0 mean.

Thanks for reading, please correct me if I am wrong.

dataset generation

you can said that you can use VoxCeleb2 to generate data i want to know that can you send overview of data set generation or how you can generate data set.

AUC calculation

Hello,

I have thoroughly reviewed your GitHub repository and noticed that there is no corresponding code available for calculating the AUC. I would like to kindly request if it would be possible for you to provide the code for calculating AUC in your repository. If the AUC code is not currently available, I would like to propose developing and contributing it via a pull request.

BA-TDF+ training goes to NaN

Dear author,

I was trying to replicate your results. I downloaded your dataset and followed the instructions presented on the GitHub page. Anyway, before reaching the end of epoch 0, around step 39K, the loss goes to NaN. I did not modify or change any of the pre-defined parameters you have put in the code. Do you have any suggestions of what could cause the issue? The same issue, is not present when I train BA-TFD.

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.