Implement strided_transformer_pose3d

README.md

@@ -472,6 +472,9 @@ The collection of pre-trained, state-of-the-art AI models.
 | [<img src="pose_estimation_3d/3dmppe_posenet/output_pose_2d.jpg" width=128px>](pose_estimation_3d/3dmppe_posenet/) |[3dmppe_posenet](/pose_estimation_3d/3dmppe_posenet/) | [PoseNet of "Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Image"](https://github.com/mks0601/3DMPPE_POSENET_RELEASE) | Pytorch | 1.2.6 and later | |
 | [<img src="pose_estimation_3d/gast/img/output_040.png" width=128px>](pose_estimation_3d/gast/) |[gast](/pose_estimation_3d/gast/) | [A Graph Attention Spatio-temporal Convolutional Networks for 3D Human Pose Estimation in Video (GAST-Net)](https://github.com/fabro66/GAST-Net-3DPoseEstimation) | Pytorch | 1.2.7 and later | [EN](https://medium.com/axinc-ai/gast-a-machine-learning-model-that-predicts-a-3d-skeleton-from-a-2d-skeleton-44449d1ff78d) [JP](https://medium.com/axinc/gast-2d%E3%81%AE%E9%AA%A8%E6%A0%BC%E3%81%8B%E3%82%893d%E3%81%AE%E9%AA%A8%E6%A0%BC%E3%82%92%E4%BA%88%E6%B8%AC%E3%81%99%E3%82%8B%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%83%A2%E3%83%87%E3%83%AB-6c70fc4b3b3a) |
 | [<img src="pose_estimation_3d/mediapipe_pose_world_landmarks/output.png" width=128px>](pose_estimation_3d/mediapipe_pose_world_landmarks/) |[mediapipe_pose_world_landmarks](/pose_estimation_3d/mediapipe_pose_world_landmarks/) | [MediaPipe Pose real-world 3D coordinates](https://google.github.io/mediapipe/solutions/pose.html#pose_world_landmarks) | TensorFlow Lite | 1.2.10 and later | |
+| [<img src="pose_estimation_3d/strided_transformer_pose3d/3D_output.png" width=128px>](pose_estimation_3d/strided_transformer_pose3d/) |[strided_transformer_pose3d](/pose_estimation_3d/strided_transformer_pose3d/) | [Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation](https://github.com/Vegetebird/StridedTransformer-Pose3D) | Pytorch | 1.2.10 and later | |
+
+
 
 ## Road detection
 

pose_estimation_3d/strided_transformer_pose3d/LICENCE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2016 Julieta Martinez, Rayat Hossain, Javier Romero
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

pose_estimation_3d/strided_transformer_pose3d/README.md

@@ -0,0 +1,57 @@
+# Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
+
+## Input
+
+![Input](input.jpg)
+
+Ailia input shape: (1, 3, 416, 416)
+
+## Output
+
+![Output 2D Pose](2D_output.png)
+
+![Output 3D Pose](3D_output.png)
+
+## Usage
+
+Automatically downloads the onnx and prototxt files on the first run.
+It is necessary to be connected to the Internet while downloading.
+
+For the sample image,
+``` bash
+$ python3 strided_transformer_pose3d.py
+```
+
+If you want to specify the input image, put the image path after the `--input` option.  
+You can use `--savepath` option to change the name of the output file to save.
+```bash
+$ python3 strided_transformer_pose3d.py --input IMAGE_PATH --savepath SAVE_IMAGE_PATH
+```
+
+By adding the `--video` option, you can input the video.   
+If you pass `0` as an argument to VIDEO_PATH, you can use the webcam input instead of the video file.
+```bash
+$ python3 strided_transformer_pose3d.py --video VIDEO_PATH
+```
+
+The default setting is to use the optimized model and weights, but you can also switch to the normal model by using the --normal option.
+
+## Reference
+
+[Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation](https://github.com/Vegetebird/StridedTransformer-Pose3D)
+
+## Framework
+
+Pytorch
+
+## Model Format
+
+ONNX opset = 12
+
+## Netron
+
+[yolox_tiny.opt.onnx.prototxt](https://netron.app/?url=https://storage.googleapis.com/ailia-models/yolox/yolox_tiny.opt.onnx.prototxt)
+
+[hrnet_w48-8ef0771d.opt.onnx.prototxt](https://netron.app/?url=https://storage.googleapis.com/ailia-models/strided_transformer_pose3d/hrnet_w48-8ef0771d.opt.onnx.prototxt)
+
+[no_refine_4365.opt.onnx.prototxt](https://netron.app/?url=https://storage.googleapis.com/ailia-models/strided_transformer_pose3d/no_refine_4365.opt.onnx.prototxt)