Andrew Zisserman

TL;DR: RepNet as discussed by the authors constrains the period prediction module to use temporal self-similarity as an intermediate representation bottleneck that allows generalization to unseen repetitions in videos in the wild, which substantially exceeds the state of the art performance on existing periodicity (PERTUBE) and repetition counting (QUVA).

TL;DR: In this article, the authors examined the information available from the motion of specularities (highlights) due to known movements by the viewer and showed that the concave/convex surface ambiguity can be resolved without knowledge of the light source position.

TL;DR: The majority of modern deep learning models are able to interpolate the data: the empirical loss can be driven near zero on all samples simultaneously and this property is exploited for the design of a new optimization algorithm for deep learning.

TL;DR: An automatic method to describe clinically useful information about scanning, and to guide image interpretation in ultrasound (US) videos of the fetal heart to achieve performance on par with expert annotations is presented.

TL;DR: The AXES project participated in the interactive instance search task (INS), the known-item search task, and the multimedia event detection task (MED) for TRECVid 2012 as mentioned in this paper.