{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.12.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"colab":{"provenance":[],"gpuType":"T4"},"accelerator":"GPU","kaggle":{"accelerator":"gpu","dataSources":[{"sourceType":"datasetVersion","sourceId":15076443,"datasetId":9652351,"databundleVersionId":15959152},{"sourceType":"datasetVersion","sourceId":15076242,"datasetId":9652215,"databundleVersionId":15958939},{"sourceType":"datasetVersion","sourceId":15078829,"datasetId":9652624,"databundleVersionId":15961779}],"dockerImageVersionId":31287,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# Install PyTorch with CUDA\n!pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126\n\n# Other required packages\n!pip install pillow matplotlib opencv-python huggingface_hub\n\n!pip install iopath fvcore\n\n!pip install hydra-core omegaconf einops\n\n!pip install ftfy\n!pip install decord","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"w7oa2AhxrVHf","outputId":"10db1a0b-a4cc-4960-ec8a-6b0f22d97c20","trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!git clone https://github.com/facebookresearch/sam3.git\n%cd sam3","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Ta2DXzU2reC1","outputId":"376c38c9-c1c3-4435-9fc4-633586b58268","trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from huggingface_hub import login\n\nlogin()","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"cZI0yCUTrii_","outputId":"0d97ab4f-b6c2-49c9-88b4-10ce96a6fd9f","trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch\nimport matplotlib.pyplot as plt\nfrom PIL import Image\nimport numpy as np\n\nfrom sam3.model_builder import build_sam3_image_model\nfrom sam3.model.sam3_image_processor import Sam3Processor","metadata":{"id":"IQOgYUddsbag","trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Load model\nmodel = build_sam3_image_model()\n\n# Processor\nprocessor = Sam3Processor(model)\n\nprint(\"SAM3 model loaded successfully\")","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":810},"id":"6xSri1DPtBiS","outputId":"a7871dd0-c377-4e94-d505-9f59084fe729","trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"image_path = \"/kaggle/input/datasets/devharis/too-complex-dataset-sam3/original.jpg\"\n\n# Option 2 (alternative): If you manually uploaded to working directory\n# image_path = \"/kaggle/working/your-image.jpg\"\n\n# Open and display image\nimage = Image.open(image_path).convert(\"RGB\")\n\nplt.imshow(image)\nplt.axis(\"off\")\nplt.title(\"Uploaded Image\")\nplt.show()","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Set image\nstate = processor.set_image(image)\n\n# Prompt\nprompt = \"motorcycle\"\n\n# Run inference\noutput = processor.set_text_prompt(\n    state=state,\n    prompt=prompt\n)\n\nmasks = output[\"masks\"]\nboxes = output[\"boxes\"]\nscores = output[\"scores\"]\n\nprint(\"Detected objects:\", len(masks))","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"image_np = np.array(image)\n\nplt.figure(figsize=(8,8))\nplt.imshow(image_np)\n\nfor mask in masks:\n    mask = mask.squeeze().cpu().numpy()   # move from GPU -> CPU -> NumPy\n    plt.imshow(mask, alpha=0.4)\n\nplt.axis(\"off\")\nplt.title(f\"SAM3 Segmentation Result, Prompt: {prompt}  \")","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}