cd /news/machine-learning/towards-seconds-level-video-represen… · home › topics › machine-learning › article
[ARTICLE · art-170308] src=arxiv.org ↗ pub= topic=machine-learning verified=true sentiment=↑ positive

Towards Seconds-level Video Representation: Quadratically Gated Feature Grid with Adaptive, Prediction-free Sampling

Researchers released Q-GAS, a coordinate-queryable implicit neural video representation that fits and quantizes video in under 50 seconds on a single RTX PRO 6000, according to the arXiv paper (2611.17786v1). Q-GAS pairs multi-resolution explicit 3D feature grids with a SwiGLU-style quadratic-gating decoder and an error-aware, prediction-free sampler; it decodes at over 370 FPS with mixed precision and about 146 FPS in FP32 on the UVG Beauty/Bosphorus/Jockey sequences. In the authors' PyTorch implementation, Q-GAS trains and quantizes 5.5-7.2 times faster than the evaluated Instant-NGP baseline and 19-28 times faster than RNeRV, and under a 90-second budget improves PSNR/MS-SSIM/LPIPS over evaluated NeRV-family pipelines at comparable BPP; code is posted at github.com/MillenRosen/Q-GAS.

by read1 min views4 publishedNov 16, 2026

arXiv:2611.17786v1 Announce Type: new Abstract: Implicit neural video representations are notoriously slow to build, which does not cross the basic practical threshold for most industrial applications. Here, we present Q-GAS, a coordinate-queryable video representation that couples multi-resolution explicit 3D feature grids with a compact SwiGLU-style quadratic-gating (QG) decoder and an error-aware, prediction-free sampler. The explicit grids avoid hash-collision coefficient sharing at the cost of modestly higher storage yet greatly boost the expressive capacity; the QG decoder introduces low-rank local second-order interactions to improve the nonlinear feature response. During fitting, a global error map periodically refreshes a high-error pool, and each batch mixes pool samples with uniform samples, which is free from the auxiliary model to predict difficult samples, thereby further escalating the efficiency. On UVG Beauty/Bosphorus/Jockey, Q-GAS fits and quantizes under 50 s on one RTX PRO 6000, decodes at over 370 FPS with mixed precision and about 146 FPS in FP32. In our PyTorch implementation, Q-GAS trains and quantizes 5.5-7.2 times faster than the evaluated Instant-NGP baseline and 19-28 times faster than RNeRV. Under a 90 s budget, it improves PSNR/MS-SSIM/LPIPS over the evaluated NeRV-family pipelines at comparable BPP over the measured overlap. We have shared our code at https://github.com/MillenRosen/Q-GAS for readers' free use.

── more in #machine-learning 4 stories · sorted by recency
── more on @q-gas 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/towards-seconds-leve…] indexed:0 read:1min 2026-11-16 · —