AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models

Shouwei Ruan, Yinpeng Dong, Liu Hanqing, Yao Huang, Hang Su, Xingxing Wei

ECCV 2024 · Vision-language and multimodal › CLIP and zero-shot · approach: Test-time adaptation of VLMs · 5 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 cross-viewpoint alignment loss L_VC on top of L_ITC and VIFormer (ablation)OVT-OpenCLIP ViT-L/14 · Viewpoint-OOD avgAvg Top-1 accuracy: 57.5 -> 62.3Improved
add VIFormer module (ablation)OVT-OpenCLIP ViT-L/14 · Viewpoint-OOD avgAvg Top-1 accuracy: 56.0 -> 57.5Improved
fine-tune with only image-text contrastive loss on multi-view data (ablation)OpenCLIP ViT-L/14 · Clean avgAvg Top-1 accuracy: 81.9 -> 79.4Hurt
add L_VC to retain clean accuracy (ablation)OVT-OpenCLIP ViT-L/14 · Clean avgAvg Top-1 accuracy: 79.9 -> 81.8Improved
minimax outlier selection instead of random outlier viewpoint sampling (ablation)OVT-OpenCLIP ViT-B/32 · Viewpoint-OOD avgAvg Top-1 accuracy: 51.5 -> 52.6Improved
apply OVT fine-tuning to CLIP ViT-B/32, B/16, L/14OpenCLIP · Viewpoint-OOD avgAvg Top-1 accuracy: +9.6%, +10.2%, +8.9%Improved
apply OVT fine-tuningOpenCLIP · 2D-OOD avgAvg Top-1 accuracy: -2.6%, -1.4%, -0.2%Hurt
use OVT-CLIP as visual encoder instead of OpenAI CLIP in LLaVA-7bLLaVA-7b · ImageNet-V+ captioningDescription accuracy beta@Adp.: 32.1 -> 41.0Improved
use adversarially trained robust CLIP (TeCoA, FARE) as LLaVA encoderLLaVA-7b · ImageNet-V+ captioningDescription accuracy beta@Adp.: 32.1 -> 22.6 (TeCoA), 23.1 (FARE)Hurt
increase lambda (weight of viewpoint consistency loss) (ablation)OVT-OpenCLIP ViT-B/32 · Clean and 2D-OOD avgAvg Top-1 accuracyMixed
use OVT-CLIP in OpenFlamingo-3BOpenFlamingo-3B · OOD-CV Pose beta@0.5VQA accuracy: 87.0 -> 82.5Mixed

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