AI Paper Atlas

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

FreestyleRet: Retrieving Images from Style-Diversified Queries

Hao Li, Yanhao Jia, Peng Jin, Zesen Cheng, Kehan Li, Jialu Sui, Chang Liu, Li Yuan

ECCV 2024 · Vision-language and multimodal › Human and embodied modeling · approach: Composed and text-based image retrieval · 9 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
apply style-init prompt tuning (FreestyleRet) to CLIP encoderCLIP · DSR Sketch->ImageR@1: 63.6 -> 80.6Improved
apply FreestyleRet to CLIP for text queriesCLIP · DSR Text->ImageR@1: 72.2 -> 69.9Hurt
prompt tune cross-modality models for style-diversified inputsCLIP · DSR Sketch->ImageR@1: 47.5 -> 63.6Improved
add auxiliary style query (sketch, art, low-res) alongside text queryFreestyleRet · DSR Sketch+Text->ImageR@1: 69.9 -> 82.5Improved
add auxiliary sketch or art query alongside text queryCLIP* and BLIP* baselines · DSR Art+Text->ImageR@1: 72.2 -> 57.8 (CLIP); 74.3 -> 58.3 (BLIP)Hurt
initialize prompts from style space (shallow) and gram matrix (deep) instead of random (ablation)FreestyleRet · DSR Sketch->ImageR@1: 78.1 -> 80.6Improved
insert style-initialized prompts in both shallow and deep layers instead of one (ablation)FreestyleRet · DSR Low-Res->ImageR@1: 84.7 -> 86.4Improved
use 4 prompt tokens instead of 1, 2 or 8 (ablation)FreestyleRet · DSRR@1Improved
apply FreestyleRet on a CLIP encoder instead of BLIP on ImageNet-X low-res queriesFreestyleRet · ImageNet-X Low-Res->ImageR@1: 97.5 (BLIP) vs 69.4 (CLIP)Hurt
add style extraction and prompt tuning on frozen encoderCLIP · Efficiencytrainable parameters and inference time: 427M -> 476M; 68ms -> 96msMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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