json

main.py

@@ -147,14 +147,16 @@ def pose2d(video, kpt_threshold):
                             thickness=4,
                             rebase_keypoint_height=True,
                             kpt_thr=kpt_threshold,
-                            device=device
+                            device=device,
+                            pred_out_dir = vis_out_dir
                             )    
     
     result = [result for result in result_generator] #next(result_generator)        
 
     out_file = glob.glob(os.path.join(vis_out_dir, "*.mp4")) #+ glob.glob(os.path.join(vis_out_dir, "*.webm")) 
-   
-    return "".join(out_file)
+    kpoints = glob.glob(os.path.join(vis_out_dir, "*.json"))
+
+    return "".join(out_file), kpoints
 
 
 def pose2dhand(video, kpt_threshold):
@@ -192,18 +194,19 @@ with block:
                         video_input = gr.Video(source="upload", type="filepath", height=612)
                         # Insert slider with kpt_thr
                         file_kpthr = gr.Slider(0, 1, value=0.3, label='Keypoint threshold')
-
-                        submit_pose_file = gr.Button("Make 2d pose estimation", variant="primary")
-                        submit_pose3d_file = gr.Button("Make 3d pose estimation", variant="primary")
-                        submit_hand_file = gr.Button("Make 2d hand estimation", variant="primary")
-                        submit_detect_file = gr.Button("Detect and track objects", variant="primary")
+                        with gr.Row():
+                            submit_pose_file = gr.Button("Make 2d pose estimation", variant="primary")
+                            submit_pose3d_file = gr.Button("Make 3d pose estimation", variant="primary")
+                            submit_hand_file = gr.Button("Make 2d hand estimation", variant="primary")
+                            submit_detect_file = gr.Button("Detect and track objects", variant="primary")
 
                 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)
                     video_output3 = gr.PlayableVideo(height=512,  label = "Estimate human hand poses", show_label=True)
                     video_output4 = gr.Video(height=512, label = "Detection and tracking", show_label=True, format="mp4")
-
+                jsonoutput = gr.Code("json")
+                
         with gr.Tab("Record video with webcam"):
             
             with gr.Column():
@@ -212,11 +215,11 @@ with block:
                         webcam_input = gr.Video(source="webcam", height=612)
                         
                         web_kpthr = gr.Slider(0, 1, value=0.3, label='Keypoint threshold')
-
-                        submit_pose_web = gr.Button("Make 2d pose estimation", variant="primary")
-                        submit_pose3d_web = gr.Button("Make 3d pose estimation", variant="primary")
-                        submit_hand_web = gr.Button("Make 2d hand estimation", variant="primary")
-                        submit_detect_web = gr.Button("Detect and track objects", variant="primary")
+                        with gr.Row():
+                            submit_pose_web = gr.Button("Make 2d pose estimation", variant="primary")
+                            submit_pose3d_web = gr.Button("Make 3d pose estimation", variant="primary")
+                            submit_hand_web = gr.Button("Make 2d hand estimation", variant="primary")
+                            submit_detect_web = gr.Button("Detect and track objects", variant="primary")
                 with gr.Row():
                     webcam_output1 = gr.PlayableVideo(height=716,  label = "Estimate human 2d poses", show_label=True)
                     webcam_output2 = gr.PlayableVideo(height=716,  label = "Estimate human 3d poses", show_label=True)
@@ -259,7 +262,7 @@ with block:
         # From file
         submit_pose_file.click(fn=pose2d, 
                             inputs=  [video_input, file_kpthr], 
-                            outputs = video_output1,
+                            outputs = [video_output1, jsonoutput],
                             queue=False)
         
         submit_pose3d_file.click(fn=pose3d,