dihuangdh / hhor Goto Github PK
View Code? Open in Web Editor NEWReconstructing Hand-Held Objects from Monocular Video, Siggraph Asia 2022 Conference Track
License: Apache License 2.0
hhor's People
Contributors
Stargazers
Watchers
hhor's Issues
ModuleNotFoundError: No module named 'easymocap'
Hi @dihuangdh
I am trying to run the demo you provided
When I run "sh apps/3_background_matting.sh", I get "easymocap" missing error
More information below:
python tools/preprocess_mask.py /scratch/2/user/aswamy/projects/hhor_evaluation/HHOR/data/DEMO_OUT/3_Giuliano
Traceback (most recent call last):
File "tools/preprocess_mask.py", line 14, in
from easymocap.mytools.camera_utils import read_camera
ModuleNotFoundError: No module named 'easymocap'
Some question about the dataset
Hi, I have some confusion about the dataset and evaluation.
- What the unit of the GT object CAD model?
- Did you use the post-processing when evaluate the CD value?
Path EasyMocap/data/smplx/smplh/MANO_LEFT.pkl does not exist!
Not able to download model weights
Hi authors,
I tried downloading the model checkpoints but I couldn't because they are restricted behind University of Sydney account restrictions. Could you make them available for download by anyone? One way to test this out is to try downloading them from an incognito browser window.
Edit: I was able to download the demo data but not the pretrained models.
Thank you for your time,
Georgi
The readme link failed and the pre-trained model could not be downloaded.
Hi, this is really a perfect piece of work, I am eager to download and learn. However, the two links in the Download pretrained models section in readme are no longer available. The connection in the Running Download demo data module has the same problem, looking forward to your update, so that we can enjoy and enjoy this masterpiece, thank you again!
Questions about dense reconstruction
Hi,
Thanks for your nice work. I am tring Neus for dense reconstruction. DEMO data does not seem to have the NeuS input.
I follow the Neus Data Convention, I run into this error: "Mismatch between number of poses and images"
So I built colmap from source following the installation guide on the official, and it still doesn't seem to work. How can i get the COLMAP output with Monocular Video?
data_dir=/amax/zxyun/3D_ObjRecon/NeuS/public_data/hhor
colmap feature_extractor --database_path ${data_dir}/database.db --image_path ${data_dir}/images
colmap sequential_matcher --database_path ${data_dir}/database.db
colmap mapper --database_path ${data_dir}/databse.db --image_path ${data_dir}/images --output_path ${data_dir}/sparse
Thanks.
Query on annoatation json file
Hi @dihuangdh ,
I am trying to understand the contents of json annotation file(below). Could you please provide give a bit of information about the following keys 'bbox', 'isKeyFrame', 'bbox_handl2d', 'handl2d', 'bbox_handr2d', 'handr2d'
Thank yoU!
{
"filename": "images/video/000524.jpg",
"height": 1080,
"width": 1920,
"annots": [
{
"personID": 0,
"bbox": [
0,
0,
1,
1,
0.0
],
"isKeyframe": false,
"bbox_handl2d": [
398.9124755859375,
486.13124084472656,
1129.7125244140625,
988.5562591552734,
0.7893120050430298
],
"handl2d": [
[
459.8125,
908.625,
0.8347628712654114
],
[
586.6875,
769.0625,
0.9201062321662903
],
[
713.5625,
680.25,
0.910774290561676
],
[
802.375,
616.8125,
0.8641476035118103
],
[
903.875,
528.0,
0.8651046752929688
],
[
815.0625,
819.8125,
0.8690438866615295
],
[
954.625,
781.75,
0.7268086671829224
],
[
1018.0625,
743.6875,
0.8126951456069946
],
[
1068.8125,
680.25,
0.48706093430519104
],
[
891.1875,
883.25,
0.8874386548995972
],
[
929.25,
781.75,
0.9016233086585999
],
[
891.1875,
731.0,
0.5650331377983093
],
[
777.0,
781.75,
0.31172749400138855
],
[
878.5,
934.0,
0.9956439733505249
],
[
916.5625,
819.8125,
0.810371994972229
],
[
853.125,
756.375,
0.8294466137886047
],
[
738.9375,
819.8125,
0.6589134931564331
],
[
853.125,
946.6875,
0.9917328357696533
],
[
878.5,
870.5625,
0.9089866876602173
],
[
802.375,
819.8125,
0.7879027128219604
],
[
726.25,
819.8125,
0.6362267136573792
]
],
"bbox_handr2d": [
351.078125,
465.4624938964844,
951.265625,
1032.9125061035156,
0.4138560891151428,
0.5399186611175537,
0.4699651896953583
],
"handr2d": [
[
401.09375,
958.34375,
0.31637895107269287
],
[
601.15625,
985.625,
0.43210628628730774
],
[
710.28125,
958.34375,
0.47122740745544434
],
[
810.3125,
912.875,
0.07449894398450851
],
[
892.15625,
512.75,
0.6837643384933472
],
[
837.59375,
940.15625,
0.3948653042316437
],
[
837.59375,
803.75,
0.6793787479400635
],
[
746.65625,
803.75,
0.5877781510353088
],
[
728.46875,
867.40625,
0.5255769491195679
],
[
892.15625,
876.5,
0.43262234330177307
],
[
892.15625,
776.46875,
0.5509440898895264
],
[
801.21875,
758.28125,
0.5797207951545715
],
[
746.65625,
794.65625,
0.608152449131012
],
[
864.875,
894.6875,
0.349997878074646
],
[
901.25,
831.03125,
0.37641897797584534
],
[
828.5,
740.09375,
0.462666779756546
],
[
783.03125,
803.75,
0.23945589363574982
],
[
864.875,
912.875,
0.20677721500396729
],
[
846.6875,
821.9375,
0.3077460825443268
],
[
837.59375,
785.5625,
0.22910256683826447
],
[
810.3125,
749.1875,
0.18179935216903687
]
]
}
],
"isKeyframe": false
}
Failed to link the data set to the pre-trained model
This is an incredibly good piece of work, but unfortunately the data set and pre-trained model links in the readme seem to be broken, looking forward to an update soon so that I can experience this masterpiece.
FileNotFoundError: [Errno 2] No such file or directory: 'pytorch_resnet101.pth'
Dear @dihuangdh
At the matting step, i get file not found error.
I changed the path of this checkpoint to "work_dirs/deeplabv3plus_r101-d8_512x512_160k_handseg3/latest.pth"
Is this correct? or should I need to download "pytorch_resnet101.pth" their github page and use it?
Thanks in Advance!
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.