AI Paper Atlas

Research map / Video / Video generation

Next-Frame Decoding for Ultra-Low-Bitrate Image Compression with Video Diffusion Priors

Yunuo Chen, Chuqin Zhou, Jiangchuan Li, Xiaoyue Ling, Bing He, Jincheng Dai, Li Song, Guo Lu

ECCV 2026 · Video › Video generation · approach: Efficient video diffusion models · 3 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
add Stage I interval adaptation (ablation)NeFIC · DIV2KBD-rate LPIPS: -52.32 -> -61.71Improved
add Stage II one-step bypass (ablation)NeFIC · DIV2KBD-rate LPIPS/FID: LPIPS -47.01 -> -61.71; FID -53.72 -> -49.94Mixed
add conditional anchor encoding (ablation)NeFIC · DIV2KBD-rate LPIPS: -56.81 -> -61.71Improved
add bypass refinement (ablation)NeFIC · DIV2KBD-rate LPIPS: -48.01 -> -61.71Improved
add auxiliary anchor loss (ablation)NeFIC · DIV2KBD-rate LPIPS: -19.78 -> -61.71Improved
add CLIP-based GAN loss (ablation)NeFIC · DIV2KBD-rate FID/LPIPS: FID -30.31 -> -49.94; LPIPS -63.09 -> -61.71Mixed
raise LoRA rank/alpha from 64/64 to 256/256 (ablation)NeFIC · DIV2KBD-rate LPIPS: -44.18% -> -61.71%Improved
raise LoRA rank/alpha from 64/64 to 256/256 (ablation)NeFIC · DIV2KBD-rate FID: -59.91 -> -49.94Hurt
add Stage II one-step generation (vs 50-step denoising) (ablation)NeFIC · DIV2Kdecoding time on Kodak (s): 30.61 -> 0.88Improved
add Stage II one-step generation (vs 50-step denoising) (ablation)NeFIC · DIV2Kdecoding time on DIV2K (s): 296.02 -> 6.83Improved
use CogVideoX VDM with anchor instead of Flux token concatenation (ablation)NeFIC · DIV2KBD-rate LPIPS: -42.18 -> -61.71Improved
add anchor token concatenation to Flux-dev-12B (vs no anchor) (ablation)Flux-dev-12B codec · DIV2KBD-rate LPIPS: -32.01 -> -42.18Improved

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