cuda

Dockerfile

@@ -7,7 +7,8 @@ COPY ./requirements.txt /code/requirements.txt
 #RUN pip install torch==1.9.1 torchvision==0.10.1 -f https://download.pytorch.org/whl/torch_stable.ht -extra-index-url https://download.pytorch.org/whl/cu111
 #https://download.pytorch.org/whl/torch_stable.ht
 
-RUN pip install torch torchvision --extra-index-url https://download.pytorch.org/whl/cu113
+RUN pip install torch torchvision --index-url https://download.pytorch.org/whl/cu117
+# --extra-index-url https://download.pytorch.org/whl/cu113
 RUN pip install --no-cache-dir --upgrade -U openmim
 RUN mim install --no-cache-dir --upgrade mmengine
 RUN mim install "mmcv>=2.0.1"

README.md

@@ -8,4 +8,4 @@ pinned: false
 python_version: 3.9
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

main.py

@@ -29,7 +29,6 @@ human3d = MMPoseInferencer(pose3d="human3d")
 track_model = YOLO('yolov8n.pt')  # Load an official Detect model
 
 
-# ultraltics
 if torch.cuda.is_available():
     device = "cuda"
 else:
@@ -192,7 +191,7 @@ def run_UI():
                             submit_pose3d_file = gr.Button("Make 3d pose estimation")
                             submit_hand_file = gr.Button("Make 2d hand estimation")
                             submit_detect_file = gr.Button("Detect and track objects")
-                            
+
                     with gr.Row():
                         video_output1 = gr.PlayableVideo(height=512,  label = "Estimate human 2d poses", show_label=True)
                         video_output2 = gr.PlayableVideo(height=512,  label = "Estimate human 3d poses", show_label=True)
@@ -230,16 +229,23 @@ def run_UI():
                         show_label=False,
                     )
                 
-                gr.Markdown("""Information about the models
-                            Pose models: `mmpose` is a library for human pose estimation that provides pre-trained models for 2D and 3D pose estimation. 
-                            The 2D pose model is used for estimating the 2D coordinates of human body joints from an image or a video frame. The model uses a convolutional neural network (CNN) to predict the joint locations and their confidence scores. 
-                            The 2D hand model is a specialized version of the 2D pose model that is designed for hand pose estimation. It uses a similar CNN architecture to the 2D pose model but is trained specifically for detecting the joints in the hand. 
-                            The 3D pose model is used for estimating the 3D coordinates of human body joints from an image or a video frame. The model uses a combination of 2D pose estimation and depth estimation to infer the 3D joint locations. 
-                            All of these models are pre-trained on large datasets and can be fine-tuned on custom datasets for specific applications. 
+                gr.Markdown(""" 
+                            \n # Information about the models 
                             
-                            Ultralight detection and tracking model: The `track()` method in the Ultralight model is used for object tracking in videos. It takes a video file or a camera stream as input and returns the tracked objects in each frame. The method uses the COCO dataset classes for object detection and tracking. The COCO dataset contains 80 classes of objects such as person, car, bicycle, etc. See https://docs.ultralytics.com/datasets/detect/coco/ for all available classes. The `track()` method uses the COCO classes to detect and track the objects in the video frames.
-                            The tracked objects are represented as bounding boxes with labels indicating the class of the object. The Ultralight model is designed to be fast and efficient, making it suitable for real-time object tracking applications.""")
+                            \n ## Pose models: All the pose estimation models comes from the library [MMpose](https://github.com/open-mmlab/mmpose). It is a library for human pose estimation that provides pre-trained models for 2D and 3D pose estimation. 
 
+                            \n ### The 2D pose model is used for estimating the 2D coordinates of human body joints from an image or a video frame. The model uses a convolutional neural network (CNN) to predict the joint locations and their confidence scores. 
+                            
+                            \n ### The 2D hand model is a specialized version of the 2D pose model that is designed for hand pose estimation. It uses a similar CNN architecture to the 2D pose model but is trained specifically for detecting the joints in the hand. 
+                            
+                            \n ### The 3D pose model is used for estimating the 3D coordinates of human body joints from an image or a video frame. The model uses a combination of 2D pose estimation and depth estimation to infer the 3D joint locations. 
+                            
+                            \n ### All of these models are pre-trained on large datasets and can be fine-tuned on custom datasets for specific applications. 
+                            
+                            \n ## Ultralight detection and tracking model: The `track()` method in the Ultralight model is used for object tracking in videos. It takes a video file or a camera stream as input and returns the tracked objects in each frame. The method uses the COCO dataset classes for object detection and tracking. 
+                            
+                            \n ### The COCO dataset contains 80 classes of objects such as person, car, bicycle, etc. See https://docs.ultralytics.com/datasets/detect/coco/ for all available classes. The `track()` method uses the COCO classes to detect and track the objects in the video frames.
+                            The tracked objects are represented as bounding boxes with labels indicating the class of the object.""")
         
         # From file
         submit_pose_file.click(fn=pose2d,