feat: update Python processors and add utility scripts

- Update ASR, face, OCR, pose processors - Add release pre-flight check script - Add synonym generation, chunk processing scripts - Add face recognition, stamp search utilities
2026-04-30 15:07:49 +08:00
parent f4697396e4
commit 8f05a7c188
256 changed files with 60505 additions and 299 deletions
--- a/scripts/quick_stamp_search.py
+++ b/scripts/quick_stamp_search.py
@@ -0,0 +1,93 @@
+#!/opt/homebrew/bin/python3.11
+"""
+Quick stamp search on 20 critical frames using OWL-ViT
+"""
+
+import os
+import cv2
+import json
+import glob
+from PIL import Image
+import torch
+from transformers import OwlViTProcessor, OwlViTForObjectDetection
+
+BASE_DIR = "output/384b0ff44aaaa1f1/critical_scenes"
+RESULTS_DIR = "output/384b0ff44aaaa1f1/critical_results"
+os.makedirs(RESULTS_DIR, exist_ok=True)
+
+print("🔬 Loading OWL-ViT...")
+processor = OwlViTProcessor.from_pretrained("google/owlvit-base-patch32")
+model = OwlViTForObjectDetection.from_pretrained("google/owlvit-base-patch32")
+model.eval()
+
+SEARCH_TERMS = [
+    "postage stamp",
+    "stamp on envelope",
+    "envelope",
+    "hand holding paper",
+    "document",
+]
+
+frames = sorted(glob.glob(os.path.join(BASE_DIR, "frame_*.jpg")))
+print(f"📸 Scanning {len(frames)} critical frames...")
+
+all_detections = []
+
+for frame_path in frames:
+    frame_name = os.path.basename(frame_path)
+    sec = frame_name.replace("frame_", "").replace("s.jpg", "")
+
+    image = Image.open(frame_path).convert("RGB")
+
+    for term in SEARCH_TERMS:
+        inputs = processor(text=[[term]], images=image, return_tensors="pt")
+
+        with torch.no_grad():
+            outputs = model(**inputs)
+
+        target_sizes = torch.Tensor([image.size[::-1]])
+        results = processor.post_process_object_detection(
+            outputs=outputs, target_sizes=target_sizes, threshold=0.05
+        )
+
+        for score, label, box in zip(
+            results[0]["scores"], results[0]["labels"], results[0]["boxes"]
+        ):
+            s = float(score)
+            if s > 0.08:
+                det = {
+                    "frame": frame_name,
+                    "sec": sec,
+                    "term": term,
+                    "score": s,
+                    "bbox": box.tolist(),
+                }
+                all_detections.append(det)
+                print(f"  📍 {sec}s | {term} | {s:.2f} | bbox={box.tolist()}")
+
+                # Save crop
+                x1, y1, x2, y2 = map(int, box.tolist())
+                img = cv2.imread(frame_path)
+                crop = img[y1:y2, x1:x2]
+                if crop.size > 0:
+                    crop_name = f"stamp_{sec}s_{term.replace(' ', '_')}.jpg"
+                    cv2.imwrite(os.path.join(RESULTS_DIR, crop_name), crop)
+
+                    # Annotate
+                    cv2.rectangle(img, (x1, y1), (x2, y2), (0, 255, 0), 3)
+                    cv2.putText(
+                        img,
+                        f"{term} {s:.2f}",
+                        (x1, y1 - 10),
+                        cv2.FONT_HERSHEY_SIMPLEX,
+                        0.7,
+                        (0, 255, 0),
+                        2,
+                    )
+                    ann_name = f"annotated_{sec}s.jpg"
+                    cv2.imwrite(os.path.join(RESULTS_DIR, ann_name), img)
+
+with open(os.path.join(RESULTS_DIR, "results.json"), "w") as f:
+    json.dump(all_detections, f, indent=2)
+
+print(f"\n🏁 Found {len(all_detections)} detections. Check {RESULTS_DIR}")