Upload 3 files

.gitattributes

@@ -33,3 +33,4 @@ 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
+example.mp4 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,90 @@
+import gradio as gr
+import cv2
+import numpy as np
+
+EXAMPLE_VIDEO = "example.mp4"  
+EXAMPLE_IMAGE = "target (2).png"  
+
+def process_video_and_image(video_path, target_image_path):
+    try:
+
+        target_img = cv2.imread(target_image_path)
+        target_gray = cv2.cvtColor(target_img, cv2.COLOR_BGR2GRAY)
+
+        cap = cv2.VideoCapture(video_path)
+
+        sift = cv2.SIFT_create()
+        keypoints_target, descriptors_target = sift.detectAndCompute(target_gray, None)
+        flann = cv2.FlannBasedMatcher(dict(algorithm=1, trees=5), dict(checks=50))
+
+        output_frames = []
+
+        while cap.isOpened():
+            ret, frame = cap.read()
+            if not ret:
+                break
+
+            frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+            keypoints_frame, descriptors_frame = sift.detectAndCompute(frame_gray, None)
+
+            if descriptors_frame is not None:
+                matches = flann.knnMatch(descriptors_target, descriptors_frame, k=2)
+                good_matches = [m for m, n in matches if m.distance < 0.70 * n.distance]
+
+                if len(good_matches) > 10:
+                    src_pts = [keypoints_target[m.queryIdx].pt for m in good_matches]
+                    dst_pts = [keypoints_frame[m.trainIdx].pt for m in good_matches]
+
+                    src_pts = np.float32(src_pts).reshape(-1, 1, 2)
+                    dst_pts = np.float32(dst_pts).reshape(-1, 1, 2)
+                    matrix, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC, 5.0)
+
+                    if matrix is not None:
+                        h, w = target_gray.shape
+                        pts = np.float32([[0, 0], [0, h], [w, h], [w, 0]]).reshape(-1, 1, 2)
+                        dst = cv2.perspectiveTransform(pts, matrix)
+                        frame = cv2.polylines(frame, [np.int32(dst)], isClosed=True, color=(0, 255, 0), thickness=3)
+
+                    match_frame = cv2.drawMatches(target_img, keypoints_target, frame, keypoints_frame, good_matches, None)
+                    output_frames.append(cv2.resize(match_frame, (1200, 600)))
+
+        cap.release()
+
+        height, width, _ = output_frames[0].shape
+        out_video_path = "output_video.avi"
+        out = cv2.VideoWriter(out_video_path, cv2.VideoWriter_fourcc(*'XVID'), 10, (width, height))
+
+        for frame in output_frames:
+            out.write(frame)
+        out.release()
+
+        return out_video_path
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+with gr.Blocks() as demo:
+    gr.Markdown("## Video and Target Image Matcher")
+    
+    with gr.Row():
+        video_input = gr.File(label="Upload Video File", file_types=[".mp4", ".avi", ".mov"])
+        image_input = gr.File(label="Upload Target Image", file_types=[".png", ".jpg", ".jpeg"])
+        example_button = gr.Button("Process Example")
+
+    output_video = gr.Video(label="Matched Video Output")
+    process_button = gr.Button("Process")
+
+    def process_example():
+        return process_video_and_image(EXAMPLE_VIDEO, EXAMPLE_IMAGE)
+    
+    process_button.click(
+        fn=process_video_and_image,
+        inputs=[video_input, image_input],
+        outputs=output_video
+    )
+    example_button.click(
+        fn=process_example,
+        inputs=[],
+        outputs=output_video
+    )
+
+demo.launch()

example.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec88256467e2bd1d07f74b1958fc689538c4b0aae442502fa0a33bb587fba2a2
+size 29809959