5 papers
MARCO: Navigating the Unseen Space of Semantic Correspondence
Claudia Cuttano, Gabriele Trivigno, Carlo Masone +1
Recent advances in semantic correspondence rely on dual-encoder architectures, combining DINOv2 with diffusion backbones. While accurate, these billion-parameter models generalize…
INSID3: Training-Free In-Context Segmentation with DINOv3
Claudia Cuttano, Gabriele Trivigno, Christoph Reich +3
In-context segmentation (ICS) aims to segment arbitrary concepts, e.g., objects, parts, or personalized instances, given one annotated visual examples. Existing work relies on (i)…
SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation
Claudia Cuttano, Gabriele Trivigno, Giuseppe Averta +1
Few-shot segmentation aims to segment unseen object categories from just a handful of annotated examples. This requires mechanisms that can both identify semantically related objec…
To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition
Davide Sferrazza, Gabriele Berton, Gabriele Trivigno +1
Visual Place Recognition (VPR) is a critical task in computer vision, traditionally enhanced by re-ranking retrieval results with image matching. However, recent advancements in VP…
SAMWISE: Infusing Wisdom in SAM2 for Text-Driven Video Segmentation
Claudia Cuttano, Gabriele Trivigno, Gabriele Rosi +2
Referring Video Object Segmentation (RVOS) relies on natural language expressions to segment an object in a video clip. Existing methods restrict reasoning either to independent sh…