Reproduce “Mask-Attention-Free Transformer for 3D Instance Segmentation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Xin Lai, Yuhui Yuan, Ruihang Chu, Yukang Chen, Han Hu, Jiaya Jia

MAFT replaces hard mask-gated attention in 3D instance segmentation with learned position queries that guide attention and are refined toward object centers. On ScanNetv2 validation, it reaches 58.4 mAP versus SPFormer’s 56.3; after 128 epochs, it reaches 57.3, compared with the baseline’s 56.9 after 512. These comparisons do not isolate center regression’s effect or establish lower wall-clock training time.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.