Josef Kittler

MICTracker: Modality Informativeness Controlled Multi-modal Fusion for Adaptive RGB-D Visual Tracking

Probabilistic Embeddings With Evidence Learning and Refinement for Text–Video Retrieval

This paper studies the problem of text-video retrieval, where the goal is to learn accurate cross-modal alignment between videos and text. This problem is challenging because of the matching ambiguity caused by the inherent gap between the heterogeneous video and text modalities. In particular, the differences in the i...

A Knowledge-Imparting Generative Modelling Framework for Heterogeneous Federated Learning

Federated learning aims to provide security for client data privacy in practical machine learning applications. In principle, a global server aggregates the models produced by local clients to obtain a global model. However, the server is challenged when collaborating with local clients handling non-identically distrib...

ReflexTrack: A Feedback-Driven Agent for Training-Free Referring Video Object Segmentation

It is demonstrated that prediction-level feedback substantially improves the reliability of training-free RVOS, with ReflexTrack, a training-free, feedback-driven agent that closes this loop at both spatial and temporal levels.