How to use glowskeleton/goblin-models with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained("glowskeleton/goblin-models") 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("glowskeleton/goblin-models") 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): ...
https://huggingface.co/glowskeleton/goblin-models/resolve/main/sam2-video/no_obj_embed_spatial.f32
hf download hf://glowskeleton/goblin-models/sam2-video/no_obj_embed_spatial.f32
curl -L -o no_obj_embed_spatial.f32 https://huggingface.co/glowskeleton/goblin-models/resolve/main/sam2-video/no_obj_embed_spatial.f32
