AI Paper Atlas

Research map / Video / Video segmentation and tracking

High-quality Single-model Deep Video Compression with Frame-Conv3D and Multi-frame Differential Modulation

Wenyu Sun, Chen Tang, Weigui Li, Zhuqing Yuan, Huazhong Yang, Yongpan Liu

ECCV 2020 · Video › Video segmentation and tracking · approach: Learned video compression · 15 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
replace Channel-Conv3D with Frame-Conv3D (ablation)video codec with traditional quantizer · UVG, half of frames droppedaverage PSNR (dB): 24.94 -> 26.37Improved
add differential quantizer in place of traditional quantizer (ablation)Frame-Conv3D codec · UVG, half of frames droppedaverage PSNR (dB): 26.37 -> 29.40Improved
add differential quantizer in place of traditional quantizer (ablation)Frame-Conv3D codec · validation subset, 0.25 BPP trainingPSNR (dB): >10 dBImproved
add differential quantizer in place of traditional quantizer (ablation)Frame-Conv3D codec · entropy of bit streams, bit width 16entropy coding gain (%): nearly 30%Improved
use 2nd order instead of 1st order differential modulation (ablation)differential quantizer · UVGPSNR and MS-SSIMNo clear effect
add prediction network (ablation)video codec · UVGPSNR and MS-SSIMImproved
add prediction loss d(F, F-bar) with weight 0.05 (ablation)video codec · UVGPSNRImproved
code 4 frames instead of 8 (ablation)video codec · UVG validation subset, 0.25 BPPconvergence and final PSNRNo clear effect
down-scale to 1/4 (M=2) instead of 1/8 (M=3) (ablation)video codec · UVG validation subset, 0.25 BPPPSNRImproved
add dropout scheme during training with 50%, 75%, 87.5% frame dropping (ablation)differential quantizer codec · UVG, variable BPPPSNR at different bit ratesImproved
compare proposed codec with H.264 and H.265 at low BPPproposed codec · UVG, HEVC Class D, Class EPSNR and MS-SSIMImproved
evaluate on fast-motion sequencesproposed codec · HEVC Class B and CPSNR and MS-SSIMHurt

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