GenRANSAC: RANSAC with AI in-loop optimization for robust geometric estimation
ducha_aiki · x · 2026-09-09
Researcher duchaaiki shared GenRANSAC, which brings AI into the inner loop of the classic RANSAC robust estimation framework, using learned components to assist sampling and hypothesis verification. Relevant for geometric computer vision work like feature matching and pose estimation; project link included in the original post.
More from Research
- Threshold ML-DSA reference code had a state-reuse flaw that could leak secret shares — jedisct1 · 2026-09-09
- Microsoft's EigenLI Compresses ColBERT Representations, Beats MUVERA Baseline — _reachsumit · 2026-09-09
- PayPal's Bottom-Up Clustering Builds Better Semantic IDs Than RQ-VAE — _reachsumit · 2026-09-09
- ByteDance Deploys SequenceO1 at Full Traffic on Douyin, Compressing 100K-Long User Histories for Recommendation — _reachsumit · 2026-09-09
- ReDSI: First Open-Source DSI Implementation Covering All Three Document Identifier Types — _reachsumit · 2026-09-09
- Perplexity Releases Q2D-Web: 190M-Doc Benchmark for Agentic RAG Retrieval — _reachsumit · 2026-09-09