AI Paper Atlas

Research map / Video / Video generation

Bitrate-Controlled Diffusion for Disentangling Motion and Content in Video

Xiao Li, Qi Chen, Xiulian Peng, Kai Yu, Xie Chen, Yan Lu

ICCV 2025 · Video › Video generation · approach: Efficient video diffusion models · 2 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 cross-driven training strategy (ablation)BCD · LRS3 motion transfercross error (x10^-2): 40.4 -> 3.67Improved
add cross-driven training strategy (ablation)BCD · LRS3 motion transferFID: 120.1 -> 86.0Improved
add bitrate loss (ablation)BCD · LRS3 motion transfercross error (x10^-2): 6.21 -> 3.67Improved
add bitrate loss (ablation)BCD · LRS3 motion transferFID: 92.7 -> 86.0Improved
lower target bitrate from 4.0 to 2.0 kbps (ablation)BCD · LRS3 motion transfercross error (x10^-2): 3.67 -> 5.68Hurt
raise target bitrate from 4.0 to 6.0 kbps (ablation)BCD · LRS3 motion transfercross error (x10^-2): 3.67 -> 4.13Hurt
raise target bitrate from 4.0 to 8.0 kbps (ablation)BCD · LRS3 motion transfercross error (x10^-2): 3.67 -> 3.74Mixed
use a single content reference frame instead of all frames (ablation)BCD · LRS3 motion transfercross error (x10^-2): 3.67 -> 3.81Hurt
use 5 content frames instead of all (ablation)BCD · LRS3 motion transfercross error (x10^-2): 3.67 -> 4.21Hurt
compare BCD with FOMM on motion errorBCD vs FOMM · LRS3 motion transfermotion error (x10^-2): 24.3 -> 3.13Improved
compare BCD with LivePortrait on FIDBCD vs LivePortrait · LRS3 motion transferFID: 100.3 -> 86.0Improved
compare BCD with C-DSVAE on attribute accuracyBCD vs C-DSVAE · Spritesattribute accuracy: 100.00 -> 100.00No clear effect

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