Hexels: JHU's Hexagonal Primitives Cut VRAM 5.5x While Beating 3DGS by 4.96dB
ssh4net · x · 2026-10-29
Researchers at Johns Hopkins CCVL introduce Hexels, planar hexagonal surface primitives with a first-order spatial color field for efficient novel view synthesis, 3D reconstruction and 2D image representation.
Key points:
- Hexagonal support is area- and boundary-efficient; the local linear color field lets each primitive cover larger coherent regions with smoothly varying appearance
- At equal model memory (19.3 MB), Hexels gain 4.96 dB PSNR over 2DGS with 5.51x lower peak VRAM (0.88 vs 4.88 GiB)
- Outperform Mip-NeRF360, 3DGS, 2DGS and other splatting methods in mean PSNR across 9 Mip-NeRF360 scenes
- On the Image-GS 2D benchmark (10K primitives) it preserves finer boundaries with lower error
Paper, project page and MIT-licensed code are public.
Related event: JHU's Hexels: Hexagonal Primitives Cut Memory and Beat Gaussian Splatting(2 posts)→
More from Research
- OpenAI's math argument still unverified: no expert has publicly assessed it — maier_ak · 2026-10-29
- OpenAI's Lean-formalized conjectures await expert review; firm calls it raw data — maier_ak · 2026-10-29
- OpenAI model's Lean-verified manuscript targets local smoothing and Kakeya conjectures — maier_ak · 2026-10-29
- OpenAI model claims 59-page Lean proof of local smoothing and Kakeya in every dimension — maier_ak · 2026-10-29
- TTME beats test-time training on math discovery with 84% fewer FLOPs — rohanpaul_ai · 2026-10-29
- Pushing one Qwen3-14B neuron turns the model into an 'Earth-Protecting Hero' — Sauers_ · 2026-10-29