AI Paper Atlas

Research map / Detection and segmentation / CNN features

LiFT: A Surprisingly Simple Lightweight Feature Transform for Dense ViT Descriptors

Saksham Suri, Matthew Walmer, Kamal Gupta, Abhinav Shrivastava

ECCV 2024 · Detection and segmentation › CNN features · approach: Plain and high-resolution ViTs · 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
add LiFT feature transform to DINO featuresDINO ViT-S/16 · SPair-71k (224 res)PCK@0.1: 24.76 -> 28.68Improved
add LiFT to DINO instead of using a larger ViT-B backboneDINO ViT-S/16 · SPair-71k (224 res)PCK@0.1: 28.68 vs 24.90 (ViT-B/16)Improved
add LiFT to DINODINO ViT-S/16 · DAVIS (224 res)J&F mean: 33.0 -> 44.3Improved
add LiFT to DINODINO ViT-S/16 · COCO20K object discovery (56 res)CorLoc: 16.28 -> 27.72Improved
add LiFT to ViTDet Mask R-CNNViTDet(MR) · COCO detectionAP: 39.50 -> 45.98Improved
add LiFT to ViTDet Cascade R-CNNViTDet(CR) · COCO segmentationAP: 40.93 -> 41.42Improved
train LiFT with self-supervision versus random initialization (ablation)DINO+LiFT · SPair-71k (224 res)PCK@0.1: 4.21 -> 28.68Improved
add image input to LiFT (ablation)DINO+LiFT · SPair-71k (224 res)PCK@0.05: 13.28 -> 14.72Improved
use cosine distance rather than L1 or L2 reconstruction loss (ablation)DINO+LiFT · SPair-71k (224 res)PCK@0.1: 27.77 (L1), 28.17 (L2) -> 28.68Improved
train LiFT longer (5 to 100 epochs) (ablation)DINO+LiFT · SPair-71k (224 res)PCK@0.1: 28.45 -> 29.11No clear effect
apply LiFT recursively (2xLiFT) (ablation)DINO+LiFT · SPair-71k (448 res)PCK@0.1: 28.68 -> 29.45 at 224; 31.38 -> 31.35 at 448Mixed
apply LiFT trained on a different backbone (ablation)DINO+LiFT · SPair-71k (448 res)PCK@0.1: 28.6 -> 9.86 (MoCo-trained LiFT on DINO would be 16.02)Hurt

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