Instructions to use PBatch23888/birdspotter-sam21-openvino with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sam2
How to use PBatch23888/birdspotter-sam21-openvino with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained(PBatch23888/birdspotter-sam21-openvino) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained(PBatch23888/birdspotter-sam21-openvino) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>): # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
BirdSpotter SAM21 OpenVINO
OpenVINO conversion of Meta's SAM 2.1 Hiera Large image encoder and box-prompted mask predictor, exported at a fixed 512 x 512 input size for BirdSpotter.
Files
openvino-512/sam21_image_encoder.xmland.bin: image encoder IR.openvino-512/sam21_mask_predictor.xmland.bin: prompt and mask decoder IR.manifest.json: SHA-256 checksums and sizes for every packaged artifact.
These models use their normal exported precision; they are not INT8 quantized. See the BirdSpotter repository for preprocessing and box-prompted inference code.
Attribution
This conversion is derived from Meta's
facebook/sam2.1-hiera-large
checkpoint. Use is subject to the upstream model's license and acceptable-use
terms.