6 papers · 1 filter
Removing Cost Volumes from Optical Flow Estimators
Simon Kiefhaber, Stefan Roth, Simone Schaub-Meyer
Cost volumes are used in every modern optical flow estimator, but due to their computational and space complexity, they are often a limiting factor regarding both processing speed…
Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery
Xinrui Gong, Oliver Hahn, Christoph Reich +4
Unsupervised multi-object discovery (MOD) aims to detect and localize distinct object instances in visual scenes without any form of human supervision. Recent approaches leverage o…
ART: Adaptive Relation Tuning for Generalized Relation Prediction
Gopika Sudhakaran, Hikaru Shindo, Patrick Schramowski +3
Visual relation detection (VRD) is the task of identifying the relationships between objects in a scene. VRD models trained solely on relation detection data struggle to generalize…
Efficient Masked Attention Transformer for Few-Shot Classification and Segmentation
Dustin Carrión-Ojeda, Stefan Roth, Simone Schaub-Meyer
Few-shot classification and segmentation (FS-CS) focuses on jointly performing multi-label classification and multi-class segmentation using few annotated examples. Although the cu…
Disentangling Polysemantic Channels in Convolutional Neural Networks
Robin Hesse, Jonas Fischer, Simone Schaub-Meyer +1
Mechanistic interpretability is concerned with analyzing individual components in a (convolutional) neural network (CNN) and how they form larger circuits representing decision mec…
Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model
Jannik Endres, Oliver Hahn, Charles Corbière +3
Omnidirectional depth perception is essential for mobile robotics applications that require scene understanding across a full 360° field of view. Camera-based setups offer a cost-e…