gsethi2409 / tf-pose-estimation Goto Github PK

I am experiencing difficulties running the run_webcam.py. This is the error I am getting, how could I resolve the issue?
Traceback (most recent call last):
File "run_webcam.py", line 8, in
from tf_pose.estimator import TfPoseEstimator
File "...\myWorkspace\tf-pose-estimation\tf_pose_init_.py", line 5, in
from tf_pose.runner import infer, Estimator, get_estimator
File "...\myWorkspace\tf-pose-estimation\tf_pose\runner.py", line 8, in
from tf_pose import eval
File "...\myWorkspace\tf-pose-estimation\tf_pose\eval.py", line 13, in
from tf_pose.estimator import TfPoseEstimator
File "...\myWorkspace\tf-pose-estimation\tf_pose\estimator.py", line 14, in
from tensorflow.python.compiler.tensorrt import trt_convert as trt
ImportError: cannot import name 'trt_convert' from 'tensorflow.python.compiler.tensorrt'

building '_pafprocess' extension
swigging pafprocess.i to pafprocess_wrap.cpp
C:\Users\Andres Paredes\Documents\anaconda3\envs\Felipe\Library\bin\swig.exe -python -c++ -o pafprocess_wrap.cpp pafprocess.i
error: Unable to find vcvarsall.bat

Hi! When I tried this, I notice that my CPU resource is all high meanwhile my GPU only used by browser, meaning that current code running only on CPU. Any idea how can we make this using GPU whenever available on the host? Thanks in advance

im having this issue. I can not get anything to run.

File "F:\SDXX\tfopenpose\openpose\tf_pose\pafprocess\pafprocess.py", line 13, in
from . import _pafprocess
ImportError: cannot import name '_pafprocess' from 'tf_pose.pafprocess'

I'm trying to run run_webcam.py file with python3 run_webcam.py --model=mobilenet_thin --resize=432x368 --camera=1
however I get this error message "AttributeError: 'NoneType' object has no attribute 'shape' "

I upgraded OpenCV and TensorFlow but it didn't work.

Thank you so much!

I tried with every possible version of numpy.
Still gets the same error

DeprecationWarning: The distutils package is deprecated and slated for removal in Python 3.12. Use setuptools or check PEP 632 for potential alternatives
from distutils.core import setup, Extension
Traceback (most recent call last):
File "C:\Users\Akshat\Desktop\CV\tf-pose-estimation\tf_pose\pafprocess\setup.py", line 15, in
include_dirs=[numpy.get_include(), "."],

AttributeError: module 'numpy' has no attribute 'get_include'

When running the train.py I get the following error:

[2020-05-31 12:31:35,782] [train] [INFO] define model+ Traceback (most recent call last): File "train.py", line 63, in <module> input_node = tf.placeholder(tf.float32, shape=(args.batchsize, args.input_height, args.input_width, 3), name='image') AttributeError: module 'tensorflow' has no attribute 'placeholder'

Using Conda and TF 2.0
Please help

When I execute run_webcam.py, memory accumulates and eventually causes a memory leak.

I think the cause is that the graph is created for each loop, but I don't know the solution.

What should I do?

KeyError: "The name 'TfPoseEstimator/image:0' refers to a Tensor which does not exist. The operation, 'TfPoseEstimator/image', does not exist in the graph."

After I change 'TfPoseEstimator/image:0' to 'image:0', new error occured.
TypeError: An op outside of the function building code is being passed
a "Graph" tensor. It is possible to have Graph tensors
leak out of the function building context by including a
tf.init_scope in your function building code.
For example, the following function will fail:
@tf.function
def has_init_scope():
my_constant = tf.constant(1.)
with tf.init_scope():
added = my_constant * 2
The graph tensor has name: Openpose/concat_stage7:0

How to resolve the below error:

git : The term 'git' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling of the
name, or if a path was included, verify that the path is correct and try again. At line:1 char:1

• git clone https://github.com/gitpython-developers/GitPython

•   + CategoryInfo          : ObjectNotFound: (git:String) [], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException
  

tf_slim is used in the source code but is not listed in requirements.txt

when i run this project like python run.py --model=mobilenet_thin --resize=432x368 --image=./images/p1.jpg
will have this problem , but I have actually installed tensorflow , i run another test project is ok

my ts version is GPU

i try to write 'python' and import tensorflow in cmd , then had a same problem ,
i try to write 'python' and import tensorflow in anaconda prompt , then activate tensorflow , enter conda list , the tensorflow is in the list , finally import tensorflow , it reminded " Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found " and "Ignore above cudart dlerror if you do not have a GPU set up on your machine."

How to solve this problem that has troubled me for a long time ？ please!
I’m just a novice, I don’t know much about it.

Hi.
When I run this command from within pafprocess directory:
swig -python -c++ pafprocess.i && python3 setup.py build_ext --inplace

I get this error:
in setup.py", line 2, in
import numpy
ModuleNotFoundError: No module named 'numpy'

There is a file called numpy.i in this directory, and I also have numpy installed in my environment.
So I am wondering what causes this error.

Thanks for your support.
Mohammad

Traceback (most recent call last): File "s5_test.py", line 59, in <module> from utils.lib_openpose import SkeletonDetector File "C:\Users\Hank\Downloads\action1\src/..\utils\lib_openpose.py", line 21, in <module> from tf_pose.networks import get_graph_path, model_wh File "C:\Users\Hank\Downloads\action1\utils/../src/githubs/tf-pose-estimation\tf_pose\__init__.py", line 5, in <module> from tf_pose.runner import infer, Estimator, get_estimator File "C:\Users\Hank\Downloads\action1\utils/../src/githubs/tf-pose-estimation\tf_pose\runner.py", line 7, in <module> from tf_pose import common File "C:\Users\Hank\Downloads\action1\utils/../src/githubs/tf-pose-estimation\tf_pose\common.py", line 10, in <module> activation_fn = tf.nn.relu AttributeError: module 'tensorflow' has no attribute 'nn'

Above message is my error while I run s5_test.py.
Did anyone get the same wrong?

2020-08-01 03:45:09.105836: F .\tensorflow/core/kernels/conv_2d_gpu.h:710] Non-OK-status: GpuLaunchKernel( SwapDimension1And2InTensor3UsingTiles<T, NumThreads, TileLongSide, TileShortSide>, total_tiles_count, NumThreads, 0, d.stream(), input, input_dims, output) status: Internal: no kernel image is available for execution on the device

This Error shows every time I try to use Tensorflow-GPU. (Tensorflow-CPU works fine but only with 2.4 FPS)

Heres the whole log file (https://pastebin.com/ccHET10x)

Any Ideas on how to fix this

My setup:

OS: Win 10 Pro 64bit

GPU: Nvidia GTX 750 Ti

Compute Capability: 5.0

Driver Version: 451.67

Cuda Toolkit Version: 10.1 u2

Cudnn Version: 7.6.5.32 for Cuda 10.1

TensorRT Version: 6.0.1.5

I'm facing an error while i run
!python run.py --model=mobilenet_thin --resize=432x368 --image=./images/p1.jpg

error is

Traceback (most recent call last):
  File "run.py", line 39, in <module>
    e = TfPoseEstimator(get_graph_path(args.model), target_size=(w, h))
  File "/content/tf-pose-estimation/tf_pose/estimator.py", line 337, in __init__
    self.tensor_image = self.graph.get_tensor_by_name('TfPoseEstimator/image:0')
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3902, in get_tensor_by_name
    return self.as_graph_element(name, allow_tensor=True, allow_operation=False)
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3726, in as_graph_element
    return self._as_graph_element_locked(obj, allow_tensor, allow_operation)
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3768, in _as_graph_element_locked
    "graph." % (repr(name), repr(op_name)))
KeyError: "The name 'TfPoseEstimator/image:0' refers to a Tensor which does not exist. The operation, 'TfPoseEstimator/image', does not exist in the graph."

I've followed the instructions provided for all the dependencies, but using pip instead of conda. I've received the following error while running the pose estimation on an image.

Type of Issue: Help required

AttributeError: module 'tensorflow._api.v1.initializers' has no attribute 'GlorotUniform'

Has anyone faced this issue or does anyone have any insights to what might be happening?

Executed Command : python run.py --model=mobilenet_thin --resize=432x368 --image=./images/p1.jpg

Operating system (Ubuntu): Ubuntu 18.04

Do you know how I could improve the detection of body parts on a 1920x1088 image with people looking pretty small on it ? Thanks

Do you know were I could find the graph_opt.pb file that should be in this directory ?