AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

See & Sniff: Learning Visuo-Olfactory Representations

Seongyu Kim, Seungwoo Lee, Hyeonggon Ryu, Joon Son Chung, Arda Senocak

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 1 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
use dense local alignment instead of global CLS-token alignment (ablation)See & Sniff · SmellNet-V-Source localizationmIoU: 0.5016 -> 0.6099Improved
add lightweight aligners to local alignment (ablation)Ours-Local · SmellNet-V-Source localizationmAP: 0.7970 -> 0.8362Improved
add lightweight aligners to local alignment (ablation)Ours-Local · SmellNet-V-Test Vision->SmellRecall@1: 55.40 -> 63.20Improved
dense local alignment vs global alignment on retrieval (ablation)See & Sniff vs Ours-Global · SmellNet-V-Test Smell->VisionRecall@10: 67.50 -> 63.90Mixed
joint visuo-olfactory model instead of cascaded SmellNet + SAM3See & Sniff · SmellNet-V-Source localizationmIoU: 0.3214 -> 0.6456Improved
use smell-conditioned alignment instead of vision-only DINOv3 attentionSee & Sniff · SmellNet-V-Source localizationmIoU: 0.5157 -> 0.6456Improved
CLIP-style global contrastive training with frozen CLIP-LargeGlobal-CLIP · SmellNet-V-Source localizationmIoU: 0.2073 vs full square 0.2072Hurt
add visual supervision (SmellNet-V) to olfactory encoder trainingScentFormer · SmellNet-Test Smell->SmellRecall@1: 52.54 -> 69.44Improved
add visual supervision to smell encoderScentFormer · SmellNet-Test classificationaccuracyImproved
visual supervision effect on per-ingredient classificationSee & Sniff · SmellNet-Test classificationaccuracy: apple 4.55 -> 77.27; avocado 61.90 -> 28.57Mixed

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