Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localization

About

Unsupervised localization and segmentation are long-standing computer vision challenges that involve decomposing an image into semantically-meaningful segments without any labeled data. These tasks are particularly interesting in an unsupervised setting due to the difficulty and cost of obtaining dense image annotations, but existing unsupervised approaches struggle with complex scenes containing multiple objects. Differently from existing methods, which are purely based on deep learning, we take inspiration from traditional spectral segmentation methods by reframing image decomposition as a graph partitioning problem. Specifically, we examine the eigenvectors of the Laplacian of a feature affinity matrix from self-supervised networks. We find that these eigenvectors already decompose an image into meaningful segments, and can be readily used to localize objects in a scene. Furthermore, by clustering the features associated with these segments across a dataset, we can obtain well-delineated, nameable regions, i.e. semantic segmentations. Experiments on complex datasets (Pascal VOC, MS-COCO) demonstrate that our simple spectral method outperforms the state-of-the-art in unsupervised localization and segmentation by a significant margin. Furthermore, our method can be readily used for a variety of complex image editing tasks, such as background removal and compositing.

Luke Melas-Kyriazi, Christian Rupprecht, Iro Laina, Andrea Vedaldi• 2022

Related benchmarks

TaskDatasetResultRank
Semantic segmentationPASCAL VOC 2012 (test)
mIoU37.2
1415
Salient Object DetectionDUTS (test)--
325
Interactive SegmentationBerkeley
NoC@907.75
235
Interactive SegmentationGrabCut
NoC@904.64
225
Camouflaged Object DetectionCOD10K (test)
S-measure (S_alpha)0.5938
224
Salient Object DetectionECSSD--
222
Interactive SegmentationDAVIS
NoC@9010.11
202
Interactive SegmentationSBD
NoC @ 90% Target11.57
171
Camouflaged Object DetectionCAMO (test)
E_phi0.6889
111
Semantic segmentationCOCO Stuff-27 (val)
mIoU890
75
Showing 10 of 50 rows

Other info

Follow for update