AI Paper Atlas

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

Open-vocabulary Object Detection via Vision and Language Knowledge Distillation

Xiuye Gu, Tsung-Yi Lin, Weicheng Kuo, Yin Cui

ICLR 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary object detection · 276 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 image-embedding distillation to text-embedding classification (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 10.1 -> 16.1Improved
Use CLIP joint image-text embeddings instead of GloVe text embeddings (ablation)ViLD-text on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 3.0 -> 10.1Improved
Ensemble ViLD-text and ViLD-image predictions using separate heads (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 16.1 -> 16.6Improved
Use ALIGN instead of CLIP as the teacher (ablation)ViLD-ensemble on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 16.6 -> 26.3Improved
Add proposal objectness to CLIP confidence scores (ablation)CLIP on cropped regions · LVIS v1 novel categoriesmask AP_r: 13.0 -> 18.9Improved
Use a single prompt instead of ensembling multiple templates and synonyms (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 15.7 -> 16.1Hurt
Add L1 image-embedding distillation to the model without distillation (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 10.4 -> 12.9; 10.4 -> 14.0; 10.4 -> 16.3; 10.4 -> 17.3Improved
Use L1 rather than L2 for image-embedding distillation (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: L1: 12.9, 14.0, 16.3, 17.3; L2: 13.7, 12.4, 13.4Improved
Increase the L1 distillation weight (ablation)ViLD on LVIS v1 · LVIS v1 novel categoriesmask AP_r: 12.9 -> 14.0 -> 16.3 -> 17.3Improved
Use image-embedding distillation alongside text classification (ablation)ViLD on COCO · COCO novel categoriesAP: 5.9 -> 27.6Improved
Transfer the LVIS-trained detector without finetuning rather than finetune (ablation)ViLD on PASCAL VOC 2007 test · PASCAL VOC 2007 testbox AP_50: 78.9 vs 72.2Hurt
Transfer the LVIS-trained detector without finetuning rather than finetune (ablation)ViLD on COCO validation · COCO validationbox AP: 39.1 vs 36.6Hurt

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