AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

LogicSeg: Parsing Visual Semantics with Neural Logic Learning and Reasoning

Liulei Li, Wenguan Wang, Yi Yang

ICCV 2023 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 49 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 logic-induced training with C, D and E rule losses (ablation)DeepLabV3+ ResNet-101 · Mapillary Vistas 2.0 val (level 1)mIoU: 37.43 -> 40.72Improved
add composition rule loss L_C alone (ablation)DeepLabV3+ ResNet-101 · Mapillary Vistas 2.0 val (level 1)mIoU: 37.43 -> 38.71Improved
add decomposition rule loss L_D alone (ablation)DeepLabV3+ ResNet-101 · Mapillary Vistas 2.0 val (level 1)mIoU: 37.43 -> 38.98Improved
add exclusion rule loss L_E alone (ablation)DeepLabV3+ ResNet-101 · Mapillary Vistas 2.0 val (level 1)mIoU: 37.43 -> 38.43Improved
add logic-induced inference with 2 message passing iterations (ablation)LogicSeg DeepLabV3+ · Mapillary Vistas 2.0 val (level 1)mIoU: 40.18 -> 40.72Improved
increase message passing iterations from 2 to 4 (ablation)LogicSeg inference · Mapillary Vistas 2.0 val (level 1)mIoU: 40.72 -> 40.03Hurt
vary aggregation coefficient q from 5 to 10 or 1 (ablation)forall quantifier approximation · Mapillary Vistas 2.0 val (level 1)mIoU: 40.72 -> 38.25 (q=10), 38.77 (q=1)Hurt
apply LogicSeg to DeepLabV3+ and MaskFormer across datasetssegmentation models · Mapillary, Cityscapes, Pascal-Part-108, ADE20KmIoU: +1.12% to +3.29% mIoUImproved
add logic-induced trainingDeepLabV3+ · Mapillary Vistas 2.0 training speedtraining time (min/epoch): 45.62 -> 47.67 (about 5%)Hurt
add logic-induced inferenceDeepLabV3+ · Mapillary Vistas 2.0 inference speedfps: 3.44 -> 3.31Hurt
LogicSeg on Cityscapes with Swin-S MaskFormer vs Hssn and SegFormerMaskFormer · Cityscapes val (level 1)mIoU: 83.85 vs SegFormer 83.90, Hssn 83.74No 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