Papers › Hypercorrelation Squeeze for Few-Shot Segmentation

Hypercorrelation Squeeze for Few-Shot Segmentation

4 Apr 2021arXiv:2104.01538archive 2025-07-28

Juhong Min, Dahyun Kang, Minsu Cho

Few-shot semantic segmentation aims at learning to segment a target object from a query image using only a few annotated support images of the target class. This challenging task requires to understand diverse levels of visual cues and analyze fine-grained correspondence relations between the query and the support images. To address the problem, we propose Hypercorrelation Squeeze Networks (HSNet) that leverages multi-level feature correlation and efficient 4D convolutions. It extracts diverse features from different levels of intermediate convolutional layers and constructs a collection of 4D correlation tensors, i.e., hypercorrelations. Using efficient center-pivot 4D convolutions in a pyramidal architecture, the method gradually squeezes high-level semantic and low-level geometric cues of the hypercorrelation into precise segmentation masks in coarse-to-fine manner. The significant performance improvements on standard few-shot segmentation benchmarks of PASCAL-5i, COCO-20i, and FSS-1000 verify the efficacy of the proposed method.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

juhongm999/hsnet officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Feature CorrelationFew-Shot Semantic SegmentationSegmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-101) FB-IoU 69.1 #60 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-101) Mean IoU 41.2 #60 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-101) learnable parameters (million) 2.5 #60 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-50) FB-IoU 68.2 #64 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-50) Mean IoU 39.2 #64 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (1-shot) HSNet (ResNet-50) learnable parameters (million) 2.5 #64 of 85 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-101) FB-IoU 72.4 #40 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-101) Mean IoU 49.5 #40 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-101) learnable parameters (million) 2.5 #40 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-50) FB-IoU 70.7 #56 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-50) Mean IoU 46.9 #56 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation COCO-20i (5-shot) HSNet (ResNet-50) learnable parameters (million) 2.5 #56 of 81 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) HSNet (ResNet-101) Mean IoU 86.5 #17 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) HSNet (ResNet-50) Mean IoU 85.5 #21 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (1-shot) HSNet (VGG-16) Mean IoU 82.3 #23 of 24 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) HSNet (ResNet-101) Mean IoU 88.5 #15 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) HSNet (ResNet-50) Mean IoU 87.8 #18 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation FSS-1000 (5-shot) HSNet (VGG-16) Mean IoU 85.8 #20 of 22 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-101) FB-IoU 77.6 #39 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-101) Mean IoU 66.2 #39 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-101) learnable parameters (million) 2.5 #39 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-50) FB-IoU 76.7 #63 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-50) Mean IoU 64.0 #63 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (ResNet-50) learnable parameters (million) 2.5 #63 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (VGG-16) FB-IoU 73.4 #83 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (1-Shot) HSNet (VGG-16) Mean IoU 59.7 #83 of 105 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-101) FB-IoU 80.6 #37 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-101) Mean IoU 70.4 #37 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-101) learnable parameters (million) 2.5 #37 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-50) FB-IoU 80.6 #45 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-50) Mean IoU 69.5 #45 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (ResNet-50) learnable parameters (million) 2.5 #45 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (VGG-16) FB-IoU 76.6 #77 of 96 Archive leaderboard report
Few-Shot Semantic Segmentation PASCAL-5i (5-Shot) HSNet (VGG-16) Mean IoU 64.1 #77 of 96 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: CP conv

CP conv

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections