ARROW unifies 3D reconstruction and point tracking from arbitrary image sets, sets new SOTA
CSProfKGD · x · 2026-10-04
Researchers released ARROW, a feed-forward model extending D4RT-style decoding to broader input types including multi-view videos and unordered image collections.
- Core innovation: an order-invariant querying approach that associates queries with observations across arbitrary viewpoints, capture times, and visibility changes
- Training on more diverse input sets improves performance and enables generalization to tasks like multi-view tracking
- Sets a new state of the art in 3D tracking on WorldTrack and TAPVid-3D, outperforms dedicated multi-view trackers on an adapted RGB-only MVTracker benchmark, and stays competitive on 3D reconstruction
- Code and weights are publicly available; paper on arXiv:2610.01314
Related event: ARROW Unifies 3D Reconstruction and Tracking from Arbitrary Images(2 posts)→
More from Research
- Paper at NeurIPS: individual parameters in weight-sparse transformers appear interpretable — CatAstro_Piyush · 2026-10-04
- How vector databases work: embeddings, cosine similarity, HNSW, IVF and PQ explained — blaizedsouza · 2026-10-04
- CoFlow adds contrastive trajectory repulsion to Flow Matching, improving few-step generation on ImageNet — burny_tech · 2026-10-04
- Latent communication papers: KV-cache integrity vs malicious agents and state-delta hybrid channels — burny_tech · 2026-10-04
- DNA design trick: fit with gradient descent first, then snap to manufacturer catalogue — anshulkundaje · 2026-10-04
- Nature study finds synaptic plasticity in biological neural networks approximates backpropagation — aran_nayebi · 2026-10-04