AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Continual Test-Time Adaptation for Single Image Defocus Deblurring via Causal Siamese Networks

Shuang Cui, Yi Li, Jiangmeng Li, Xiongxin Tang, Bing Su, Fanjiang Xu, Hui Xiong

IJCV 2025 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 6 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 CauSiam to source SIDD modelDPDNet-S · avg of 4 SIDD test setsPSNR: 24.676 -> 25.412Improved
add CauSiam to source SIDD modelNRKNet · avg of 4 SIDD test setsPSNR: 26.283 -> 26.560Improved
add CoTTA to source SIDD modelDPDNet-S · avg of 4 SIDD test setsPSNR: 24.676 -> 22.863Hurt
add TENT-continual (entropy minimization)DPDNet-S · avg of 4 SIDD test setsPSNR: 24.676 -> 18.690Hurt
add CoTTA over three roundsDRBNet · long-term CTTAPSNR: 27.409 -> 23.933Hurt
add CauSiam over three roundsDRBNet · long-term CTTAPSNR: 27.409 -> 27.648Improved
add VLM-guided semantic priors (VSPI) (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 25.077 -> 25.412Improved
add spatial consistency loss L_s (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 24.910 -> 25.412Improved
add high-frequency consistency loss L_f (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 25.384 -> 25.412Improved
update only the CA module instead of all online-model parameters (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 25.146 -> 25.412Improved
add EMA update of offline model (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 25.323 -> 25.412Improved
use DINOv2 instead of CLIP for priors (ablation)CauSiam on DPDNet-S · avg of 4 SIDD test setsPSNR: 25.412 -> 25.077Hurt

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