AI Paper Atlas

Research map / Video / Video segmentation and tracking

HiNeRV: Video Compression with Hierarchical Encoding-based Neural Representation

Ho Man Kwan, Ge Gao, Fan Zhang, Andy Gower, David Bull

NeurIPS 2023 · Video › Video segmentation and tracking · approach: Neural video representations · 22 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use HiNeRV instead of prior NeRV-style modelsvideo representation · UVG avg, scale SPSNR (dB): 32.68 -> 35.27Improved
HiNeRV vs faster NeRV baselinesBunny scale S · Bunnydecoding FPS: 202.3 -> 76.7Hurt
add bilinear upsampling with hierarchical encoding instead of sub-pixel conv 1x1 (ablation)HiNeRV · UVG ablationavg PSNR (dB): 34.78 -> 36.27Improved
use bilinear upsampling instead of sub-pixel conv 3x3 (ablation)HiNeRV · UVG ablationavg PSNR (dB): 33.70 -> 36.27Improved
add hierarchical upsampling encoding (ablation)HiNeRV · UVG ablationavg PSNR (dB): 35.95 -> 36.27Improved
hierarchical encoding instead of Fourier / local Fourier / local grid encoding (ablation)HiNeRV · UVG ablationavg PSNR (dB): 35.92/35.77/35.98 -> 36.27Improved
ConvNeXt block instead of MLP block (ablation)HiNeRV · UVG ablationavg PSNR (dB): 34.69 -> 36.27Improved
ConvNeXt block instead of two conv layers (ablation)HiNeRV · UVG ablationavg PSNR (dB): 35.12 -> 36.27Improved
unified frame/patch representation instead of frame-wise (ablation)HiNeRV · UVG ablationavg PSNR (dB): 36.10 -> 36.27Improved
unified representation instead of patch-wise (ablation)HiNeRV · UVG ablationavg PSNR (dB): 35.38 -> 36.27Improved
bilinear instead of nearest neighbor interpolation (ablation)HiNeRV · UVG ablationavg PSNR (dB): 36.08 -> 36.27Improved
HiNeRV vs conventional/learned codecs on PSNR BD-ratevideo compression · UVGBD-rate PSNR: beats x265, DCVC, VCT; HM RA 7.54% worse anchorMixed

Later papers that compared with this method

2 later accepted papers report HiNeRV as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy