Papers › Rotation equivariant vector field networks

Rotation equivariant vector field networks

29 Dec 2016ICCV 2017 10arXiv:1612.09346archive 2025-07-28

Diego Marcos, Michele Volpi, Nikos Komodakis, Devis Tuia

In many computer vision tasks, we expect a particular behavior of the output with respect to rotations of the input image. If this relationship is explicitly encoded, instead of treated as any other variation, the complexity of the problem is decreased, leading to a reduction in the size of the required model. In this paper, we propose the Rotation Equivariant Vector Field Networks (RotEqNet), a Convolutional Neural Network (CNN) architecture encoding rotation equivariance, invariance and covariance. Each convolutional filter is applied at multiple orientations and returns a vector field representing magnitude and angle of the highest scoring orientation at every spatial location. We develop a modified convolution operator relying on this representation to obtain deep architectures. We test RotEqNet on several problems requiring different responses with respect to the inputs' rotation: image classification, biomedical image segmentation, orientation estimation and patch matching. In all cases, we show that RotEqNet offers extremely compact models in terms of number of parameters and provides results in line to those of networks orders of magnitude larger.

PaperPDFConference PDFCode

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

Code

di-marcos/RotEqNet officialmentioned in paperpytorch report
COGMAR/RotEqNet mentioned on GitHubpytorch report
ojas97/roteqnetp3 mentioned 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

Breast Tumour ClassificationColorectal Gland Segmentation:Image ClassificationImage SegmentationMulti-tissue Nucleus SegmentationPatch MatchingSemantic Segmentationimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Breast Tumour Classification PCam VF-CNN (C12) AUC 0.898 #13 of 16 Archive leaderboard report
Breast Tumour Classification PCam VF-CNN (C8) AUC 0.881 #14 of 16 Archive leaderboard report
Breast Tumour Classification PCam VF-CNN (C4) AUC 0.871 #15 of 16 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C12) Dice 0.782 #12 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C12) F1-score 0.776 #12 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C12) Hausdorff Distance (mm) 251.9 #12 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C8) Dice 0.758 #13 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C8) F1-score 0.745 #13 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C8) Hausdorff Distance (mm) 287.5 #13 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C4) Dice 0.721 #14 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C4) F1-score 0.711 #14 of 15 Archive leaderboard report
Colorectal Gland Segmentation: CRAG VF-CNN (C4) Hausdorff Distance (mm) 318.9 #14 of 15 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C12) Dice 0.813 #8 of 18 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C12) Hausdorff Distance (mm) 51.4 #8 of 18 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C12) Dice 0.808 #11 of 18 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C12) Hausdorff Distance (mm) 50.7 #11 of 18 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C4) Dice 0.800 #12 of 18 Archive leaderboard report
Multi-tissue Nucleus Segmentation Kumar VF-CNN (C4) Hausdorff Distance (mm) 49.9 #12 of 18 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

Convolution

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