Welcome to the CALVIN research group homepage. We were part of the IPAB institute of the University of Edinburgh. The group closed at the end of 2018. This website still has all the contents of our research.
We researched several topics related to learning in computer vision, with focus on weakly supervised learning of object classes, semantic segmentation, and large-scale annotation.
Vittorio Ferrari moved to Google. Check this website for his research at Google.
- The CALVIN research group no longer exists. All materials remain available online at our new site calvin-vision.net
- IJCV paper on learning single-image 3D reconstruction by generative modelling of shape, pose and shading
- Software release: Differentiable Renderer for TensorFlow (DIRT)
- Paper on generating and reconstructing 3D meshes with only 2D supervision accepted to BMVC 2018
- Two papers accepted to CVPR 2018
- Technical report on generative modelling of 3D object layouts in apartments
- Software Release for Joint learning of object and action detectors
- Interview with Vicky Kalogeiton in ICCV Daily
- Software Release for Action Tubelet detector
- Paper on semantic part emergence in CNNs accepted to IJCV 2017
- Release of YouTube-Objects v2.3.
- Interview with Vitto Ferrari in CVPR Daily
- Five papers accepted to ICCV 2017.