AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Improving Image Recognition by Retrieving From Web-Scale Image-Text Data

Ahmet Iscen, Alireza Fathi, Cordelia Schmid

CVPR 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Composed and text-based image retrieval · 20 citations (OpenAlex)

Read the paper

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
Learn attention weights for retrieved examples instead of averaging their valuesRetrieval-augmented classifier · ImageNet-LTaccuracy (All): 75.8 -> 78.3Improved
Learn attention weights for retrieved examples instead of averaging their valuesRetrieval-augmented classifier · Places-LTaccuracy (All): 44.9 -> 49.9Improved
Learn attention weights for retrieved examples instead of averaging their valuesRetrieval-augmented classifier · iNaturalist2021-Miniaccuracy: 58.9 -> 66.2Improved
Learn attention weights for retrieved examples instead of averaging their valuesRetrieval-augmented classifier · Webvisionaccuracy: 78.2 -> 83.6Improved
Fine-tune the visual encoder along with the memory attention module (ablation)MAM · ImageNet-LTaccuracy (All): 78.3 -> 82.3Improved
Fine-tune the visual encoder along with the memory attention module (ablation)MAM · Places-LTaccuracy (All): 49.9 -> 51.4Improved
Use T5-Base text embeddings rather than visual memory-value embeddings (ablation)MAM · ImageNet-LTaccuracyImproved
Increase the memory dataset size (ablation)MAM · ImageNet-LTaccuracyImproved
Increase the number of retrieved neighbors on a small memory such as ImageNet-LT (ablation)MAM with T5-Base memory values · ImageNet-LTaccuracyHurt
Increase the number of retrieved neighbors on a large memory dataset (ablation)MAM with T5-Base memory values · YFCC, LAION, and WebLIaccuracyNo clear effect
Add more image-text pairs to the memory after training without retraining MAM (ablation)MAM · ImageNet-LTaccuracyImproved

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