updated files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/videos/vid_1.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/videos/vid_2.mp4 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,149 @@
+from ultralytics import YOLO
+import cv2
+import gradio as gr
+import numpy as np
+import os
+import torch
+from image_segmenter import ImageSegmenter
+
+# params
+CANCEL_PROCESSING = False
+
+img_seg = ImageSegmenter(model_type="yolov8m-seg-custom")
+
+def resize(image):
+    """
+    resize the input nd array
+    """
+    h, w = image.shape[:2]
+    if h > w:
+        return cv2.resize(image, (480, 640)) 
+    else:
+        return cv2.resize(image, (640, 480)) 
+
+def process_image(image):
+    image = resize(image)
+    prediction, _ = img_seg.predict(image)
+    return prediction
+
+
+def process_video(vid_path=None):
+    vid_cap = cv2.VideoCapture(vid_path)
+    while vid_cap.isOpened():
+        ret, frame = vid_cap.read()        
+        if ret:
+            print("Making frame predictions ....")
+            frame = resize(frame)
+            frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+            prediction, _ = img_seg.predict(frame)
+            yield prediction
+    
+    return None
+
+def update_segmentation_options(options):
+    img_seg.is_show_bounding_boxes = True if 'Show Boundary Box' in options else False
+    img_seg.is_show_segmentation = True if 'Show Segmentation Region' in options else False
+    img_seg.is_show_segmentation_boundary = True if 'Show Segmentation Boundary' in options else False
+
+def update_confidence_threshold(thres_val):
+    img_seg.confidence_threshold = thres_val/100
+
+def model_selector(model_type):
+
+    if "Small - Better performance and less accuracy" == model_type:
+        yolo_model = "yolov8s_seg_custom"
+    elif "Medium - Balanced performance and accuracy" == model_type:
+        yolo_model = "yolov8m-seg-custom"
+    elif "Large - Slow performance and high accuracy" == model_type:
+        yolo_model = "yolov8m-seg-custom"
+    else:
+        yolo_model = "yolov8m-seg-custom"
+
+    img_seg = ImageSegmenter(model_type=yolo_model)
+
+def cancel():
+    CANCEL_PROCESSING = True
+
+if __name__ == "__main__":
+
+    # gradio gui app
+    with gr.Blocks() as my_app:
+
+        # title
+        gr.Markdown("<h1><center>Hand detection and segmentation</center></h1>")
+
+        # tabs
+        with gr.Tab("Image"):
+            with gr.Row():
+                with gr.Column(scale=1):
+                    img_input = gr.Image()
+                    model_type_img = gr.Dropdown(
+                        ["Small - Better performance and less accuracy", 
+                         "Medium - Balanced performance and accuracy", 
+                         "Large - Slow performance and high accuracy"], 
+                        label="Model Type", value="Medium - Balanced performance and accuracy",
+                        info="Select the inference model before running predictions!")
+                    options_checkbox_img = gr.CheckboxGroup(["Show Boundary Box", "Show Segmentation Region"], label="Options")
+                    conf_thres_img = gr.Slider(1, 100, value=60, label="Confidence Threshold", info="Choose the threshold above which objects should be detected")
+                    submit_btn_img = gr.Button(value="Predict")                    
+
+                with gr.Column(scale=2):
+                    with gr.Row():
+                        img_output = gr.Image(height=300, label="Segmentation")
+            
+            gr.Markdown("## Sample Images")
+            gr.Examples(
+                examples=[os.path.join(os.path.dirname(__file__), "assets/images/img_1.jpg"),
+                          os.path.join(os.path.dirname(__file__), "assets/images/img_1.jpg")],
+                inputs=img_input,
+                outputs=img_output,
+                fn=process_image,
+                cache_examples=True,
+            )
+
+        with gr.Tab("Video"):
+            with gr.Row():
+                with gr.Column(scale=1):
+                    vid_input = gr.Video()
+                    model_type_vid = gr.Dropdown(
+                        ["Small - Better performance and less accuracy", 
+                         "Medium - Balanced performance and accuracy", 
+                         "Large - Slow performance and high accuracy"], 
+                        label="Model Type", value="Medium - Balanced performance and accuracy",
+                        info="Select the inference model before running predictions!")
+                    
+                    options_checkbox_vid = gr.CheckboxGroup(["Show Boundary Box", "Show Segmentation Region"], label="Options")
+                    conf_thres_vid = gr.Slider(1, 100, value=60, label="Confidence Threshold", info="Choose the threshold above which objects should be detected")
+                    with gr.Row():
+                        cancel_btn = gr.Button(value="Cancel")
+                        submit_btn_vid = gr.Button(value="Predict")
+            
+                with gr.Column(scale=2):
+                    with gr.Row():
+                        vid_output = gr.Image(height=300, label="Segmentation")
+            
+            gr.Markdown("## Sample Videos")
+            gr.Examples(
+                examples=[os.path.join(os.path.dirname(__file__), "assets/videos/vid_1.mp4"),
+                          os.path.join(os.path.dirname(__file__), "assets/videos/vid_2.mp4"),],
+                inputs=vid_input,
+                # outputs=vid_output,
+                # fn=vid_segmenation,
+            )
+            
+
+        # image tab logic
+        submit_btn_img.click(process_image, inputs=img_input, outputs=img_output)
+        options_checkbox_img.change(update_segmentation_options, options_checkbox_img, [])
+        conf_thres_img.change(update_confidence_threshold, conf_thres_img, [])
+        model_type_img.change(model_selector, model_type_img, [])
+
+        # video tab logic
+        submit_btn_vid.click(process_video, inputs=vid_input, outputs=vid_output)
+        model_type_vid.change(model_selector, model_type_vid, [])
+        cancel_btn.click(cancel, inputs=[], outputs=[])
+        options_checkbox_vid.change(update_segmentation_options, options_checkbox_vid, [])
+        conf_thres_vid.change(update_confidence_threshold, conf_thres_vid, [])       
+
+
+    my_app.queue(concurrency_count=5, max_size=20).launch(debug=True)

assets/videos/vid_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce4d3c0dc28cea27fdedbc54f9de88a57cf56981ffe3a31984631341375880af
+size 4663136

assets/videos/vid_2.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d041fda27e5fa33e46260e212fc61ed58c6155dd80299b7b1f71ee99629449a5
+size 5585030

datasets/hand_dataset_bbox/README.dataset.txt

@@ -0,0 +1,6 @@
+# Hand Detector > 2023-07-16 1:39pm
+https://universe.roboflow.com/vaishanth-ramaraj-uub2b/hand-detector-pjtzx
+
+Provided by a Roboflow user
+License: Public Domain
+

datasets/hand_dataset_bbox/README.roboflow.txt

@@ -0,0 +1,35 @@
+
+Hand Detector - v3 2023-07-16 1:39pm
+==============================
+
+This dataset was exported via roboflow.com on July 16, 2023 at 5:41 PM GMT
+
+Roboflow is an end-to-end computer vision platform that helps you
+* collaborate with your team on computer vision projects
+* collect & organize images
+* understand and search unstructured image data
+* annotate, and create datasets
+* export, train, and deploy computer vision models
+* use active learning to improve your dataset over time
+
+For state of the art Computer Vision training notebooks you can use with this dataset,
+visit https://github.com/roboflow/notebooks
+
+To find over 100k other datasets and pre-trained models, visit https://universe.roboflow.com
+
+The dataset includes 262 images.
+Hands are annotated in YOLOv8 format.
+
+The following pre-processing was applied to each image:
+* Auto-orientation of pixel data (with EXIF-orientation stripping)
+* Resize to 640x640 (Stretch)
+
+The following augmentation was applied to create 3 versions of each source image:
+* Random rotation of between -15 and +15 degrees
+* Random shear of between -15° to +15° horizontally and -15° to +15° vertically
+* Salt and pepper noise was applied to 5 percent of pixels
+
+The following transformations were applied to the bounding boxes of each image:
+* Random Gaussian blur of between 0 and 0.75 pixels
+
+

datasets/hand_dataset_bbox/data.yaml

@@ -0,0 +1,12 @@
+names:
+- hands
+nc: 1
+roboflow:
+  license: Public Domain
+  project: hand-detector-pjtzx
+  url: https://universe.roboflow.com/vaishanth-ramaraj-uub2b/hand-detector-pjtzx/dataset/3
+  version: 3
+  workspace: vaishanth-ramaraj-uub2b
+test: ../test/images
+train: /content/drive/MyDrive/My Projects/yolo_custom_training/Hand-Detector-3/train/images
+val: /content/drive/MyDrive/My Projects/yolo_custom_training/Hand-Detector-3/valid/images

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_3_mp4-15_jpg.rf.648c927a529d159315b14692b4823a73.txt

@@ -0,0 +1 @@
+0 0.37109375 0.79453125 0.128125 0.1625

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_3_mp4-24_jpg.rf.5a2c8df29ac4bd6456b94b45ff252068.txt

@@ -0,0 +1 @@
+0 0.378125 0.73203125 0.11796875 0.1578125

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_3_mp4-8_jpg.rf.b72e8e36f12ff26b13c580e247ea995b.txt

@@ -0,0 +1,2 @@
+0 0.32421875 0.44921875 0.0953125 0.16796875
+0 0.3859375 0.63984375 0.11875 0.115625

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_4_mp4-17_jpg.rf.60b53afc4553c743bf21ba016b29b551.txt

@@ -0,0 +1,2 @@
+0 0.4921875 0.409375 0.10078125 0.2375
+0 0.5078125 0.559375 0.1203125 0.178125

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_4_mp4-33_jpg.rf.9c8b9c92d4e9d64b471cf495a06b5f77.txt

@@ -0,0 +1,2 @@
+0 0.48203125 0.39140625 0.06484375 0.296875
+0 0.55625 0.3609375 0.159375 0.21015625

datasets/hand_dataset_bbox/test/labels/Sign_language_interpreter_vid_4_mp4-42_jpg.rf.d0ff24f8fa95d380f906159a39009912.txt

@@ -0,0 +1 @@
+0 0.49453125 0.21796875 0.1390625 0.28671875

datasets/hand_dataset_bbox/test/labels/sign_language_interpreter_vid_2_mp4-18_jpg.rf.2b7ded8aacb6746b672598d788785529.txt

@@ -0,0 +1,2 @@
+0 0.62265625 0.49453125 0.1796875 0.26875
+0 0.740625 0.85859375 0.16796875 0.1796875

datasets/hand_dataset_bbox/test/labels/sign_language_interpreter_vid_2_mp4-2_jpg.rf.d068568645cf6bef50ac41567ea666d2.txt

@@ -0,0 +1 @@
+0 0.5046875 0.3578125 0.15234375 0.2375

datasets/hand_dataset_bbox/test/labels/sign_language_interpreter_vid_2_mp4-6_jpg.rf.bdaa2043a4e6f5dc7845ef7f5b99371b.txt

@@ -0,0 +1 @@
+0 0.49921875 0.52578125 0.12265625 0.2265625

datasets/hand_dataset_bbox/test/labels/sign_language_interpreter_vid_2_mp4-9_jpg.rf.d258d97eb97ad74b58c993a0a6ced074.txt

@@ -0,0 +1,2 @@
+0 0.60703125 0.584375 0.19765625 0.2578125
+0 0.596875 0.8796875 0.20234375 0.22578125

datasets/hand_dataset_bbox/test/labels/sign_language_interpreter_vid_3_mp4-14_jpg.rf.de46e9b06a0ec2903988a625871f65cc.txt

@@ -0,0 +1,2 @@
+0 0.61875 0.34453125 0.1421875 0.21171875
+0 0.4828125 0.37265625 0.15703125 0.184375