AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

ESS: Learning Event-Based Semantic Segmentation from Still Images

Zhaoning Sun, Nico Messikommer, Daniel Gehrig, Davide Scaramuzza

ECCV 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive segmentation · 115 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 consistency loss in embedding space (ablation)ESS · DSEC-Semantic (UDA, no event labels)mIoU (%): 39.31 -> 44.87Improved
add consistency loss on predictions (ablation)ESS · DSEC-Semantic (UDA, no event labels)mIoU (%): 43.59 -> 44.87Improved
add consistency loss on task outputs (ablation)ESS · DSEC-Semantic (UDA, no event labels)mIoU (%): 42.78 -> 44.87Improved
add skip connections (ablation)ESS · DSEC-Semantic (UDA, no event labels)mIoU (%): 38.08 -> 44.87Improved
add embedding-space consistency loss (ablation)ESS · DSEC-Semantic (UDA, no event labels)accuracy (%): 80.86 -> 84.04Improved
ESS instead of VID2E in UDA settingESS · DDD17 (UDA, no event labels)mIoU (%): 45.48 -> 52.46Improved
ESS instead of E2VID in UDA settingESS · DSEC-Semantic (UDA, no event labels)mIoU (%): 40.70 -> 44.87Improved
add image labels to event labels in trainingESS · DSEC-Semantic (supervised)mIoU (%): 51.57 -> 53.29Improved
add image labels to event labels in trainingESS · DDD17 (supervised, event labels)mIoU (%): 61.37 -> 60.43Hurt
ESS with only event labels versus EV-SegNetESS · DSEC-Semantic (supervised)mIoU (%): 51.76 -> 51.57No clear effect
ESS instead of E2VID for embedding alignmentESS · DSEC-Semantic embedding alignment (UDA)symmetric KL dissimilarity: 0.073 -> 0.025Improved

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