AI Paper Atlas

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

Continual Adaptation for Deep Stereo

Matteo Poggi, Alessio Tonioni, Fabio Tosi, Stefano Mattoccia, Luigi Di Stefano

TPAMI 2021 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 36 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 online adaptation (FULL, photometric loss)MADNet · KITTI CityD1-all (%): 37.42 -> 3.35Improved
use SGM proxy supervision instead of photometric self-supervision (MAD)MAD · KITTI CityD1-all (%): 7.51 -> 4.12Improved
use SGM proxy supervision instead of photometric self-supervision (MAD)MAD · KITTI CampusD1-all (%): 22.27 -> 11.24Improved
use SGM proxy supervision instead of photometric loss (FULL)FULL · KITTI CampusD1-all (%): 10.62 -> 9.69Improved
use SGM proxy instead of photometric loss (FULL)FULL · KITTI CityD1-all (%): 3.35 -> 3.51No clear effect
use WILD proxy instead of photometric loss (FULL)FULL · KITTI CityD1-all (%): 3.35 -> 5.11Hurt
use LIDAR proxy instead of photometric loss (MAD)MAD · KITTI CampusD1-all (%): 22.27 -> 19.33Improved
use proxy-supervised FULL++ instead of FULLMADNet-Synthia · across domainsD1-all (%): 2.76 -> 2.23Improved
add FOL2A meta-learning to FULLMADNet-Synthia · across domainsD1-all (%): 2.76 -> 2.60Improved
use MAD (modular) instead of FULL adaptationMADNet · KITTI across domainsD1-all (%): 2.43 -> 4.09Hurt
add prior adaptation on KITTI before DrivingStereo (FULL)MADNet · DrivingStereo DuskyD1-all (%): 20.40 -> 9.21Improved
add prior adaptation on KITTI before DrivingStereo (MAD)MADNet · DrivingStereo DuskyD1-all (%): 33.23 -> 9.81Improved

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