AI Paper Atlas

Research map / Detection and segmentation / CNN features

LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models

Haiwen Huang, Anpei Chen, Volodymyr Havrylov, Andreas Geiger, Dan Zhang

ICCV 2025 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 13 citations (OpenAlex)

This paper tried a common change: Add positional encoding

add sinusoidal positional encoding and 3x3 image conv · Cityscapes segmentation (Stage 1, 50k) · mIoU: 43.13 -> 48.63 Improved

What 646 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

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
add Stage 2 self-distillation pseudo-GT on top of Stage 1 mask-bicubic (ablation)LoftUp · Cityscapes segmentationIoU: 50.43 -> 53.10Improved
use SAM mask-based pseudo-GT instead of LiFT 2x features (ablation)LoftUp · Cityscapes segmentationIoU: 35.85 -> 50.43Improved
use full-resolution pseudo-GT instead of per-image optimized features (ablation)LoftUp · COCO-Stuff segmentationIoU: 59.08 -> 61.11Improved
use coordinate-based cross-attention architecture instead of LIIF (ablation)LoftUp · COCO-Stuff segmentationmIoU: 52.24 -> 61.11Improved
apply LoftUp pseudo-GT objective to other upsamplers (ablation)resize-conv / FeatUp-JBU · COCO-Stuff segmentationmIoU: 56.05 -> 56.05 (resize-conv with LoftUp objective); FeatUp 56.30 -> 57.90Improved
add sinusoidal positional encoding and 3x3 image conv (ablation)LoftUp · Cityscapes segmentation (Stage 1, 50k)mIoU: 43.13 -> 48.63Improved
increase cross-attention blocks from 2 to 3 (ablation)LoftUp · COCO-Stuff segmentation (Stage 1)mIoU: 58.65 -> 58.35No clear effect
increase training set from 50k to 1M images (ablation)LoftUp · COCO-Stuff segmentation (Stage 1)mIoU: 58.65 -> 59.87Improved
upsample 224-input features to 448 resolution instead of 224 (ablation)LoftUp · COCO-Stuff segmentationmIoU: 61.11 -> 60.61Hurt
replace 4x larger VFM input with LoftUp upsampling at 224 inputDINOv2-S/14 · COCO-Stuff segmentationmIoU: 56.79 -> 61.11Improved
apply LiFT upsampler to VFM for interactive segmentationDINOv2-S/14 · GrabCut interactive segmentationIoU@1 click: 64.90 -> 29.66Hurt

Later papers that compared with this method

2 later accepted papers report LoftUp as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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