You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
HyperPose is a library for building human pose estimation systems that can efficiently operate in the wild.
16
+
---
6
17
7
-
> **News**: The PoseProposal inference model is released! See the HyperPose models on [Google Drive](https://drive.google.com/drive/folders/1w9EjMkrjxOmMw3Rf6fXXkiv_ge7M99jR?usp=sharing).
18
+
<palign="center">
19
+
<a href="#Features">Features</a> •
20
+
<a href="#Documentation">Documentation</a> •
21
+
<a href="#Quick-Start-with-Docker">Quick-Start with Docker</a> •
22
+
<a href="#Performance">Performance</a> •
23
+
<a href="#License">License</a>
24
+
</p>
25
+
26
+
HyperPose is a library for building human pose estimation systems that can efficiently operate in the wild.
8
27
9
28
## Features
10
29
@@ -13,42 +32,63 @@ HyperPose has two key features, which are not available in existing libraries:
13
32
-**Flexible training platform**: HyperPose provides flexible Python APIs to build many useful pose estimation models (e.g., OpenPose and PoseProposalNetwork). HyperPose users can, for example, customize data augmentation, use parallel GPUs for training, and replace deep neural networks (e.g., changing from ResNet to MobileNet), thus building models specific to their real-world scenarios.
14
33
-**High-performance pose estimation**: HyperPose achieves real-time pose estimation though a high-performance pose estimation engine. This engine implements numerous system optimizations: pipeline parallelism, model inference with TensorRT, CPU/GPU hybrid scheduling, and many others. This allows HyperPose to run 4x faster than OpenPose and 10x faster than TF-Pose.
15
34
16
-
## Get Started
35
+
## Documentation
36
+
37
+
You can install HyperPose(Python Training Library, C++ inference Library) and learn its APIs through [HyperPose Documentation](https://hyperpose.readthedocs.io/en/latest/).
17
38
18
-
You can install HyperPose and learn its APIs through [Documentation](https://hyperpose.readthedocs.io/en/latest/).
39
+
## Quick-Start with Docker
19
40
20
-
## Example
41
+
The official docker image is on [DockerHub](https://hub.docker.com/r/tensorlayer/hyperpose).
21
42
22
-
We provide an example to show human pose estimation achieved by HyperPose. You need to install CUDA Toolkit 10+, TensorRT 7+, OpenCV 3.2+ and gFlags (cmake version), and enable C++ 17 support. Once the prerequisite are met, run the following script:
43
+
Make sure you have [docker](https://docs.docker.com/get-docker/) with [nvidia-docker](https://github.com/NVIDIA/nvidia-docker) functionality installed.
44
+
45
+
> Also note that your nvidia driver should be [compatible](https://docs.nvidia.com/deploy/cuda-compatibility/index.html#support-title) with CUDA10.2.
# [Example 2](X11 server required to see the imshow window): Real-time inference.
55
+
# You may need to install X11 server locally:
56
+
# sudo apt install xorg openbox xauth
57
+
xhost +; docker run --rm --gpus all -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix tensorlayer/hyperpose --imshow
58
+
59
+
60
+
# [Example 3]: Camera + imshow window
61
+
xhost +; docker run --name pose-camera --rm --gpus all -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --device=/dev/video0:/dev/video0 tensorlayer/hyperpose --source=camera --imshow
62
+
# To quit this image, please type `docker kill pose-camera` in another terminal.
63
+
64
+
65
+
# [Dive into the image]
66
+
xhost +; docker run --rm --gpus all -it -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --device=/dev/video0:/dev/video0 --entrypoint /bin/bash tensorlayer/hyperpose
67
+
# For users that cannot access a camera or X11 server. You may also use:
68
+
# docker run --rm --gpus all -it --entrypoint /bin/bash tensorlayer/hyperpose
32
69
```
33
70
71
+
> For more details, please check [here](https://hyperpose.readthedocs.io/en/latest/markdown/quick_start/prediction.html#table-of-flags-for-hyperpose-cli).
72
+
34
73
## Performance
35
74
36
-
We compare the prediction performance of HyperPose with [OpenPose 1.6](https://github.com/CMU-Perceptual-Computing-Lab/openpose) and [TF-Pose](https://github.com/ildoonet/tf-pose-estimation). We implement the OpenPose algorithms with different configurations in HyperPose. The test-bed has Ubuntu18.04, 1070Ti GPU, Intel i7 CPU (12 logic cores). The test video is Crazy Updown Funk ([YouTube](https://www.youtube.com/watch?v=2DiQUX11YaY)). The HyperPose models (in the ONNX or Uff formats) are available [here](https://github.com/tensorlayer/pretrained-models/tree/master/models/hyperpose).
75
+
We compare the prediction performance of HyperPose with [OpenPose 1.6](https://github.com/CMU-Perceptual-Computing-Lab/openpose) and [TF-Pose](https://github.com/ildoonet/tf-pose-estimation). We implement the OpenPose algorithms with different configurations in HyperPose. The test-bed has Ubuntu18.04, 1070Ti GPU, Intel i7 CPU (12 logic cores).
0 commit comments