GitHub Repository: https://github.com/realjero/KeypointRCNN # Pose Estimation using Keypoint RCNN **`Mert Özmeral & Jerome Habanz`** ### 📝 Description This project implements Keypoint RCNN using PyTorch, trained on the COCO 2017 dataset. It includes a training script for model optimization, a test script for evaluation, and a user-friendly testing framework for real-time webcam analysis. The training script refines model parameters for accurate keypoint detection. The test script assesses model performance on validation data or custom images. The testing framework enables interactive webcam testing, showcasing the model's real-world capabilities. Together, these components empower users in pose estimation and related computer vision tasks. ### 🏚️Dataset For training, we have used the [COCO2017-Dataset](https://cocodataset.org/) - [train2017](http://images.cocodataset.org/zips/train2017.zip) - [val2017](http://images.cocodataset.org/zips/val2017.zip) - [annotations](http://images.cocodataset.org/annotations/annotations_trainval2017.zip) - [Data Format](https://cocodataset.org/#format-data) #### 📁Folder Structure ``` root ├───coco │ ├───annotations │ │ ├───person_keypoints_train2017.json │ │ └───person_keypoints_val2017.json │ ├───train2017 │ │ └───(...).jpg │ └───val2017 │ └───(...).jpg ├───e42_b8_lr0.02_m0.9.pth ... ``` ### 🔻Results ###### Google Drive: - [e42_b8_lr0.02_m0.9.pth](https://drive.google.com/file/d/1p6MW5PYiz9c8BSYb0iLd4qx1w64Be3Ww/view?usp=sharing)

🔰Usage

Make sure the required [packages](#packages) are installed!

🏋️Training

``` python train.py ``` ###### Output: ``` losses=9.61300 loss_classifier=0.76776 loss_box_reg=0.01154 loss_keypoint=8.07662 loss_objectness=0.69867 loss_rpn_box_reg=0.05841: 0%| | 0/42 [00:01

✅Testing

🎥Live Showcase with Webcam

``` python detect.py ```

🎥Showcase Image/Video

``` python detect.py --file image.jpg python detect.py --file video.mp4 ``` ###### Output:

📚Packages

The project is compatible with 🐍Python 3.10 You can find the packages used in _requirements.txt_ or down below⬇️ ``` matplotlib numpy opencv_python Pillow pycocotools torch torchvision tqdm ```

### 🐢References - https://pytorch.org/vision/main/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn - https://cocodataset.org