Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Hyperpixel Flow: Semantic Correspondence with Multi-layer Neural Features

About

Establishing visual correspondences under large intra-class variations requires analyzing images at different levels, from features linked to semantics and context to local patterns, while being invariant to instance-specific details. To tackle these challenges, we represent images by "hyperpixels" that leverage a small number of relevant features selected among early to late layers of a convolutional neural network. Taking advantage of the condensed features of hyperpixels, we develop an effective real-time matching algorithm based on Hough geometric voting. The proposed method, hyperpixel flow, sets a new state of the art on three standard benchmarks as well as a new dataset, SPair-71k, which contains a significantly larger number of image pairs than existing datasets, with more accurate and richer annotations for in-depth analysis.

Juhong Min, Jongmin Lee, Jean Ponce, Minsu Cho• 2019

Related benchmarks

TaskDatasetResultRank
Semantic CorrespondenceSPair-71k (test)
PCK@0.128.2
122
Semantic CorrespondencePF-WILLOW
PCK@0.1 (bbox)76.3
109
Semantic CorrespondencePF-Pascal (test)
PCK@0.184.8
106
Semantic CorrespondencePF-PASCAL
PCK @ alpha=0.184.8
98
Keypoint TransferSPair-71k (test)
Bicycle18.9
38
Semantic CorrespondencePF-WILLOW (test)
PCK @ 0.10 (bbox)74.4
37
Semantic keypoint transferPF-Pascal (test)
PCK @ 0.0563.5
35
Semantic CorrespondenceCaltech-101
LT-ACC88
31
Semantic CorrespondencePF-PASCAL
PCK@0.188.3
29
Semantic MatchingTSS (test)
FG3DCar PCK@0.0593.6
27
Showing 10 of 13 rows

Other info

Code

Follow for update