Andrew Zisserman

TL;DR: A learning algorithm is developed that counts the number of cells in a large field of view image automatically, and can be used to investigate colony growth in time lapse sequences, using a novel, small, and cost effective diffraction device.

TL;DR: A strong teacher network for facial emotion recognition that achieves the state of the art on a standard benchmark is developed and it is shown that the speech emotion embedding can be used for speech emotion recognition on external benchmark datasets.

TL;DR: The results demonstrated the benefits of estimating heart rate combined from multiple regions of interest using data fusion and the convolutional neural network can be used to detect the presence of a patient and segment the patient’s skin area for vital-sign estimation, thus enabling the automatic continuous monitoring of vital signs in a hospital environment.

TL;DR: In this paper, an analysis of the way in which optical elements distort the appearance of their backgrounds allows the construction of environment mattes in situ without the need for specialized calibration, which is a powerful technique for modeling the complex light-transport properties of real-world optically active elements: transparent, refractive and reflective objects.

TL;DR: A novel automatic system for Modic changes classification of vertebral endplates, which operates on T1 and T2 MRI, trained and validated using a large dataset of 785 patients, containing MRIs sourced from a wide range of acquisition protocols.