from ultralytics import YOLO

model = YOLO("beanyzoldyck/tiktok-watermark-yolo")
results = model.predict("video.mp4", conf=0.10)

for result in results:
    for box in result.boxes:
        x1, y1, x2, y2 = box.xyxy[0].tolist()
        conf = box.conf[0].item()
        print(f"Watermark detected at ({x1:.0f}, {y1:.0f}, {x2:.0f}, {y2:.0f}) conf={conf:.4f}")

Training

Base model: YOLOv11n ( yolo11n.pt )
Task: Object detection (single class — TikTok watermark)
Input size: 640×640
Format: PyTorch (.pt)

Trained with Ultralytics YOLO on a curated TikTok watermark dataset using yolo11n as the base checkpoint. Default training configuration: 640px image size.

Intended Use

Detecting TikTok watermarks for content moderation, video processing pipelines, and research purposes.