RGBD20K: new RGB-D segmentation benchmark with 20K image pairs and 160 categories
UNT · hf · 2026-09-26
UNT released RGBD20K, a new benchmark for RGB-D semantic segmentation: 160 fine-grained categories (vs NYUv2's 40, SUN RGB-D's 37), 20,000 RGB-D image pairs, and rigorously re-annotated ground truth fixing long-standing label noise. The accompanying score-purified fusion (SPF) method achieves SOTA across all evaluated benchmarks. Dataset available on GitHub (ShaohuaDong2021/RGBD20K).
More from Research
- Biopharma Bench: agents complete only 8 of 71 real biopharma tasks, GPT-6 Astra leads — AllThingsApx · 2026-09-26
- Matryoshka Attribution finds neural network circuits via gradient descent, tops interpretability benchmark by 2.9x — stanfordnlp · 2026-09-26
- Anthropic's science blog shows Claude pulling off 'Nine Loops' particle physics calculations — rohanpaul_ai · 2026-09-26
- UniReps workshop invites NeurIPS rejects, deadline Oct 4 for Paris event — Pseudomanifold · 2026-09-26
- Creative Destruction Lab opens 15-year tech startup panel dataset to researchers — avicgoldfarb · 2026-09-26
- Building the next OCR benchmark: skalskip92 seeks hard samples from handwriting to technical drawings — BLUECOW009 · 2026-09-26