xuebinqin/BASNet

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1436 stars · 249 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

BASNet is the official PyTorch implementation of the CVPR 2019 paper 'BASNet: Boundary-Aware Salient Object Detection', a deep learning model for accurately segmenting the most prominent objects in images. It also covers the extended boundary-aware segmentation network for salient objects in clutter (SOC) and camouflaged object detection (COD), with pre-trained models and training/inference scripts.

Use cases

• segment the main object in a photo with precise boundaries
• run salient object detection on images with a pretrained model
• train a boundary-aware segmentation network on my own dataset
• detect camouflaged objects in images
• generate saliency maps for benchmark datasets like DUTS or ECSSD
• extract foreground objects for image editing or background removal

When to choose

• you need research-grade salient object detection with strong boundary accuracy
• you want a PyTorch model with pretrained weights for SOD, SOC, or COD
• you are reproducing or building on the CVPR 2019 BASNet paper

When to avoid

• you need a production-ready, actively maintained segmentation library
• you want a simple API or packaged install rather than running scripts
• you need real-time inference on mobile or web (consider the newer U^2-Net instead)

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision computer-vision image-processing deep-learning machine-learning python cross-platform salient-object-detection segmentation pytorch cvpr-2019 boundary-aware camouflaged-object-detection research-code

Member repositories

For agents

markdown · JSON · MCP: product_card(name="xuebinqin/BASNet")