AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Towers of Babel: Combining Images, Language, and 3D Geometry for Learning Multimodal Vision

Xiaoshi Wu, Hadar Averbuch-Elor, Jin Sun, Noah Snavely

ICCV 2021 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 16 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 3D contrastive loss (ablation)classification baseline · WikiScenes WS-KmAP: 70.8 -> 75.3Improved
add 3D contrastive loss (ablation)classification baseline · WikiScenes WS-UmAP: 48.3 -> 52.0Improved
use inter-image negative sampling instead of 3D MSE loss (ablation)3D consistency loss · WikiScenes WS-KmAP: 71.4 -> 75.3Improved
use contrastive loss instead of 3D triplet loss (ablation)3D consistency loss · WikiScenes WS-KmAP: 72.1 -> 75.3Improved
3D MSE loss added to baseline (ablation)classification baseline · WikiScenes WS-UmAP: 48.3 -> 49.5Improved
intra-image vs inter-image negative sampling (ablation)3D contrastive loss · WikiScenes WS-UmAP: 52.5 vs 52.0No clear effect
add pixel-level self-supervised loss (ablation)semantic classification loss · WikiScenes WS-UmAP: 53.8 -> 52.0No clear effect
add pixel-level self-supervised loss (ablation)semantic classification loss · WikiScenes WS-KmAP: 73.4 -> 75.3Improved
add 3D contrastive lossResNet-50 · WikiScenes WS-KmAP: 68.5 -> 73.9Improved
add 3D contrastive lossMobileNetV2 · WikiScenes WS-UmAP: 50.2 -> 53.4Improved
add 3D contrastive loss2D segmentation model · WikiScenes segmentationprecision: 68.6 -> 80.8Improved
add 3D contrastive loss2D segmentation model · WikiScenes segmentationIoU: 25.4 -> 27.2Improved

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