AI Paper Atlas

Research map / Image generation / Image restoration

FreqSIC: Frequency-aware Stereo Image Compression with Bi-directional Checkerboard Context Model

Shiyu Qin, Yongkang Lu, Yimin Zhou, Jiawei Li, Yifan Ren, Yuerong Xue, Shu-Tao Xia, Bin Chen

CVPR 2026 · Image generation › Image restoration · approach: Learned image compression

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 FSCT instead of plain stereo cross attention (ablation)FreqSIC · InStereo2KBD-PSNR vs full model: -0.251 dB without itImproved
add frequency dynamic fusion (adaptive weighting of high/low bands) (ablation)FreqSIC FSCT · InStereo2KBD-PSNR vs full model: -0.178 dB without itImproved
use learnable pooling-based frequency decomposition instead of Haar wavelet (ablation)FreqSIC FSCT · InStereo2KBD-PSNR vs full model: -0.906 dB with HaarImproved
use FSCT instead of BiSIC mutual attention (ablation)FreqSIC · CityscapesBD-PSNR vs full model: -0.203 dB with mutual attentionImproved
add inter-view priors in the checkerboard context model (ablation)FreqSIC · InStereo2KBD-PSNR vs full model: -0.648 dB without themImproved
use bi-directional checkerboard context model instead of BCSIC context model (ablation)FreqSIC · CityscapesBD-PSNR vs full model: -0.648 dB with BCSICImproved
use checkerboard context model instead of BiSIC autoregressive context model (ablation)FreqSIC · InStereo2KBD-PSNR vs full model: -0.135 dB with BiSICImproved
replace spatial autoregressive entropy model with checkerboard context modelFreqSIC vs BiSIC · InStereo2K coding latencytotal latency (s): 78.60 -> 1.62Improved
FreqSIC versus BiSIC overallFreqSIC · InStereo2KBD-Rate (PSNR): -48.07% -> -52.43%Improved
increase target bitrate (lambda) in FDF moduleFreqSIC · InStereo2Khigh-frequency weightImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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