From the 1 of 6 linked papers with an AI index.
6 papers
SVI360: Spherical Video Interpolation
Le-Kim Nguyen, Renato Martins, Pascal Vasseur +1
The paper introduces SVI360, a dual‑branch neural framework that improves interpolation of omnidirectional (360°) videos by combining the original view with a rotated orthogonal vi…
Cross-Domain Human Action Recognition from Multiview Motion and Textual Descriptions
Yannick Porto, Renato Martins, Thomas Chalumeau +1
Robustness to domain changes is a key capability for effective deployment of human action recognition systems in real-world scenarios, where action categories at inference can pres…
Improving Viewpoint-Invariance and Temporal Consistency for Action Detection
Yannick Porto, Renato Martins, Thomas Chalumeau +1
Viewpoint change invariance and action temporal consistency are critical aspects for the effective deployment of human action detection of untrimmed videos. Existing appearance-bas…
Adverse-to-the-eXtreme Panoptic Segmentation: URVIS 2026 Study and Benchmark
Yiting Wang, Nolwenn Peyratout, Tim Brodermann +13
This paper presents the report of the URVIS 2026 challenge on adverse-to-extreme panoptic segmentation. As the first challenge of its kind, it attracted 17 registered participants…
Dense Scene Reconstruction from Light-Field Images Affected by Rolling Shutter
Hermes McGriff, Renato Martins, Nicolas Andreff +1
This paper presents a dense depth estimation approach from light-field (LF) images that is able to compensate for strong rolling shutter (RS) effects. Our method estimates RS compe…
Leveraging Semantic Cues from Foundation Vision Models for Enhanced Local Feature Correspondence
Felipe Cadar, Guilherme Potje, Renato Martins +2
Visual correspondence is a crucial step in key computer vision tasks, including camera localization, image registration, and structure from motion. The most effective techniques fo…