IEEE

LVAR-CZSL: Learning Visual Attributes Representation for Compositional Zero-Shot Learning

Abstract: Compositional Zero-Shot Learning (CZSL) has been applied to various scenarios, including scene understanding, visual-language representation, and domain adaptation. Despite numerous ...
IEEE

VCINet: Visual Cue-Inspired Feature Learning Network for SAR Building Segmentation

Abstract: The extraction of building areas from synthetic aperture radar (SAR) imagery is of great importance for urbanization assessment and emergency response. However, SAR images are generated by ...
GitHub

Official Implement of "Agentic Jigsaw Interaction Learning for Enhancing Visual Perception and Reasoning in Vision-Language Models"

# Clone the AGILE repository git clone https://github.com/yuzeng0-0/AGILE.git cd AGILE # Create a new Conda environment with Python 3.10 conda create -n agile python ...
Artnet

Painter Ronan Day-Lewis Brings His Haunting Visual World to Film With ‘Anemone’

Anemone, the new film by artist Ronan Day-Lewis, opens with a kid’s drawing. The work takes the form of a long scroll, illustrated with scenes of increasing carnage—men wielding guns in one tableau, ...
Medical Xpress

Brain narrows focus from broad visual cues to fine details, study shows

How we focus our attention before we even see an object matters. For example, when we look for something moving in the sky, our expectation would be very different if the object is a bird flying past ...
GitHub

robot-learning-freiburg/loopgnn

If you find our work useful, please consider citing our paper: @article{buechner2025visual, author={Büchner, Martin and Dahiya, Liza and Dorer, Simon and Ramtekkar, Vipul and Nishiimiya, Kenji and ...