detect-objects.py

#!/usr/bin/env python3
# /// script
# requires-python = ">=3.10"
# dependencies = [
#     "transformers@git+https://github.com/huggingface/transformers.git@1fba72361e8e0e865d569f7cd15e5aa50b41ac9a",
#     "datasets",
#     "huggingface-hub",
#     "pillow",
#     "tqdm",
#     "torchvision",
#     "accelerate",
# ]
# ///

"""
Detect objects in images using Meta's SAM3 (Segment Anything Model 3).

This script processes images from a HuggingFace dataset and detects a single object
type based on a text prompt, outputting bounding boxes in HuggingFace object detection format.

Examples:
    # Detect photographs in historical newspapers
    uv run detect-objects.py \\
        davanstrien/newspapers-with-images-after-photography \\
        my-username/newspapers-detected \\
        --class-name photograph

    # Detect animals in camera trap images
    uv run detect-objects.py \\
        wildlife-images \\
        wildlife-detected \\
        --class-name animal \\
        --confidence-threshold 0.6

    # Test on small subset
    uv run detect-objects.py input output \\
        --class-name table \\
        --max-samples 10

    # Run on HF Jobs with L4 GPU
    hf jobs uv run --flavor l4x1 \\
        -s HF_TOKEN=$HF_TOKEN \\
        https://huggingface.co/datasets/uv-scripts/sam3/raw/main/detect-objects.py \\
        input-dataset output-dataset \\
        --class-name photograph \\
        --confidence-threshold 0.5

Performance:
    - L4 GPU: ~2-4 images/sec (depending on image size and batch size)
    - Memory: ~8-12 GB VRAM
    - Recommended batch size: 4-8 for L4, 8-16 for A10

Note: To detect multiple object types, run the script multiple times with different
      --class-name values and merge the results.
"""

import argparse
import logging
import os
import sys
from typing import List, Dict, Any

import torch
from PIL import Image
from datasets import load_dataset, Dataset, Features, Sequence, Value, ClassLabel
from datasets import Image as ImageFeature
from huggingface_hub import HfApi, login
from tqdm.auto import tqdm
from transformers import Sam3Processor, Sam3Model

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s - %(levelname)s - %(message)s",
    datefmt="%H:%M:%S",
)
logger = logging.getLogger(__name__)

# GPU availability check
if not torch.cuda.is_available():
    logger.error("❌ CUDA is not available. This script requires a GPU.")
    logger.error("For local testing, ensure you have a CUDA-capable GPU.")
    logger.error("For cloud execution, use HF Jobs with --flavor l4x1 or similar.")
    sys.exit(1)


def parse_args():
    """Parse command line arguments."""
    parser = argparse.ArgumentParser(
        description="Detect objects in images using SAM3",
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog=__doc__,
    )

    # Required arguments
    parser.add_argument(
        "input_dataset", help="Input HuggingFace dataset ID (e.g., 'username/dataset')"
    )
    parser.add_argument(
        "output_dataset", help="Output HuggingFace dataset ID (e.g., 'username/output')"
    )

    # Object detection configuration
    parser.add_argument(
        "--class-name",
        required=True,
        help="Object class to detect (e.g., 'photograph', 'animal', 'table')",
    )
    parser.add_argument(
        "--confidence-threshold",
        type=float,
        default=0.5,
        help="Minimum confidence score for detections (default: 0.5)",
    )
    parser.add_argument(
        "--mask-threshold",
        type=float,
        default=0.5,
        help="Threshold for mask generation (default: 0.5)",
    )

    # Dataset configuration
    parser.add_argument(
        "--image-column",
        default="image",
        help="Name of the column containing images (default: 'image')",
    )
    parser.add_argument(
        "--split", default="train", help="Dataset split to process (default: 'train')"
    )
    parser.add_argument(
        "--max-samples",
        type=int,
        default=None,
        help="Maximum number of samples to process (for testing)",
    )
    parser.add_argument(
        "--shuffle", action="store_true", help="Shuffle dataset before processing"
    )

    # Processing configuration
    parser.add_argument(
        "--batch-size",
        type=int,
        default=4,
        help="Batch size for processing (default: 4)",
    )
    parser.add_argument(
        "--model",
        default="facebook/sam3",
        help="SAM3 model ID (default: 'facebook/sam3')",
    )
    parser.add_argument(
        "--dtype",
        default="bfloat16",
        choices=["float32", "float16", "bfloat16"],
        help="Model precision (default: 'bfloat16')",
    )

    # Output configuration
    parser.add_argument(
        "--private", action="store_true", help="Make output dataset private"
    )
    parser.add_argument(
        "--hf-token",
        default=None,
        help="HuggingFace token (default: uses HF_TOKEN env var or cached token)",
    )

    return parser.parse_args()


def load_and_validate_dataset(
    dataset_id: str,
    split: str,
    image_column: str,
    max_samples: int = None,
    shuffle: bool = False,
    hf_token: str = None,
) -> Dataset:
    """Load dataset and validate it has the required image column."""
    logger.info(f"📂 Loading dataset: {dataset_id} (split: {split})")

    try:
        dataset = load_dataset(dataset_id, split=split, token=hf_token)
    except Exception as e:
        logger.error(f"Failed to load dataset '{dataset_id}': {e}")
        sys.exit(1)

    # Validate image column exists
    if image_column not in dataset.column_names:
        logger.error(f"Column '{image_column}' not found in dataset")
        logger.error(f"Available columns: {dataset.column_names}")
        sys.exit(1)

    # Shuffle if requested
    if shuffle:
        logger.info("🔀 Shuffling dataset")
        dataset = dataset.shuffle()

    # Limit samples if requested
    if max_samples is not None:
        logger.info(f"🔢 Limiting to {max_samples} samples")
        dataset = dataset.select(range(min(max_samples, len(dataset))))

    logger.info(f"✅ Loaded {len(dataset)} samples")
    return dataset


def process_batch(
    batch: Dict[str, List[Any]],
    image_column: str,
    class_name: str,
    processor: Sam3Processor,
    model: Sam3Model,
    confidence_threshold: float,
    mask_threshold: float,
) -> Dict[str, List[List[Dict[str, Any]]]]:
    """Process a batch of images and return detections for a single class."""
    images = batch[image_column]

    # Convert to PIL Images and ensure RGB
    pil_images = []
    for img in images:
        if isinstance(img, str):
            img = Image.open(img)
        if img.mode == "L":
            img = img.convert("RGB")
        elif img.mode != "RGB":
            img = img.convert("RGB")
        pil_images.append(img)

    # Process batch through model
    try:
        inputs = processor(
            images=pil_images,
            text=[class_name] * len(pil_images),  # Same prompt for all images
            return_tensors="pt",
        ).to(model.device, dtype=model.dtype)

        with torch.no_grad():
            outputs = model(**inputs)

        # Post-process outputs using original_sizes from processor
        results = processor.post_process_instance_segmentation(
            outputs,
            threshold=confidence_threshold,
            mask_threshold=mask_threshold,
            target_sizes=inputs.get("original_sizes").tolist(),
        )

    except Exception as e:
        logger.warning(f"⚠️  Failed to process batch: {e}")
        # Return empty detections for all images in batch
        return {"objects": [[] for _ in range(len(pil_images))]}

    # Convert to HuggingFace object detection format
    batch_objects = []
    for result in results:
        boxes = result.get("boxes", torch.tensor([]))
        scores = result.get("scores", torch.tensor([]))

        # Handle empty results
        if len(boxes) == 0:
            batch_objects.append([])
            continue

        # Build list of detections
        detections = []
        for box, score in zip(boxes.cpu().numpy(), scores.cpu().numpy()):
            x1, y1, x2, y2 = box
            width = x2 - x1
            height = y2 - y1

            detection = {
                "bbox": [float(x1), float(y1), float(width), float(height)],
                "category": 0,  # Single class, always index 0
                "score": float(score),
            }
            detections.append(detection)

        batch_objects.append(detections)

    return {"objects": batch_objects}


def main():
    args = parse_args()

    class_name = args.class_name.strip()
    if not class_name:
        logger.error("❌ Invalid --class-name argument. Provide a class name.")
        sys.exit(1)

    logger.info("🚀 SAM3 Object Detection")
    logger.info(f"   Input: {args.input_dataset}")
    logger.info(f"   Output: {args.output_dataset}")
    logger.info(f"   Class: {class_name}")
    logger.info(f"   Confidence threshold: {args.confidence_threshold}")
    logger.info(f"   Batch size: {args.batch_size}")

    # Authentication
    if args.hf_token:
        login(token=args.hf_token)
    elif os.getenv("HF_TOKEN"):
        login(token=os.getenv("HF_TOKEN"))

    # Load dataset
    dataset = load_and_validate_dataset(
        args.input_dataset,
        args.split,
        args.image_column,
        args.max_samples,
        args.shuffle,
        args.hf_token,
    )

    # Load model
    logger.info(f"🤖 Loading SAM3 model: {args.model}")
    try:
        processor = Sam3Processor.from_pretrained(args.model)
        model = Sam3Model.from_pretrained(
            args.model, torch_dtype=getattr(torch, args.dtype), device_map="auto"
        )
        logger.info(f"✅ Model loaded on {model.device}")
    except Exception as e:
        logger.error(f"❌ Failed to load model: {e}")
        logger.error("Ensure the model exists and you have access permissions")
        sys.exit(1)

    # Define output schema before processing
    logger.info("📊 Creating output schema...")
    new_features = dataset.features.copy()
    new_features["objects"] = [
        {
            "bbox": [float, float, float, float],
            "category": class_name,
            "score": float,
        }
    ]

    # Process dataset with explicit output features
    logger.info("🔍 Processing images...")
    processed_dataset = dataset.map(
        lambda batch: process_batch(
            batch,
            args.image_column,
            class_name,
            processor,
            model,
            args.confidence_threshold,
            args.mask_threshold,
        ),
        batched=True,
        batch_size=args.batch_size,
        features=new_features,
        desc="Detecting objects",
    )

    # Calculate statistics
    total_detections = sum(len(objs) for objs in processed_dataset["objects"])
    images_with_detections = sum(len(objs) > 0 for objs in processed_dataset["objects"])

    logger.info("✅ Detection complete!")
    logger.info(f"   Total detections: {total_detections}")
    logger.info(
        f"   Images with detections: {images_with_detections}/{len(processed_dataset)}"
    )
    logger.info(
        f"   Average detections per image: {total_detections / len(processed_dataset):.2f}"
    )

    # Push to hub
    logger.info(f"📤 Pushing to HuggingFace Hub: {args.output_dataset}")
    try:
        processed_dataset.push_to_hub(args.output_dataset, private=args.private)
        logger.info(
            f"✅ Dataset available at: https://huggingface.co/datasets/{args.output_dataset}"
        )
    except Exception as e:
        logger.error(f"❌ Failed to push to hub: {e}")
        logger.info("💾 Saving locally as backup...")
        processed_dataset.save_to_disk("./output_dataset")
        logger.info("✅ Saved to ./output_dataset")
        sys.exit(1)


if __name__ == "__main__":
    main()