I realised I didn't update the website in quite a while, so this is just a short post to say that I have finished my PhD last year (hoooray!) and that I started a job as research scientist at a startup company based in London, called MindVisionLabs (hooooray!).

You can check out the cool and exciting things we are working on by clicking here.

Fun fun fun! :)

We uploaded our paper entitled Discriminative convolutional Fisher vector network for action recognition to arXiv. You can check out the paper by following this link.

Abstract

Discriminative convolutional Fisher vector network for action recognition

Petar Palasek, Ioannis Patras

In this work we propose a novel neural network architecture for the problem of human action recognition in videos. The proposed architecture expresses the processing steps of classical Fisher vector approaches, that is dimensionality reduction by principal component analysis (PCA) projection, Gaussian mixture model (GMM) and Fisher vector descriptor extraction, as network layers. By contrast to other methods where these steps are performed consecutively and the corresponding parameters are learned in an unsupervised manner, having them defined as a single neural network allows us to refine the whole model discriminatively in an end to end fashion. Furthermore, we show that the proposed architecture can be used as a replacement for the fully connected layers in popular convolutional networks achieving a comparable classification performance, or even significantly surpassing the performance of similar architectures while reducing the total number of trainable parameters by a factor of 5. We show that our method achieves significant improvements in comparison to the classical chain.

Our work on Deep globally constrained MRFs for Human Pose Estimation was accepted at the International Conference on Computer Vision (ICCV 2017) which will be held from October 22nd to 29th, 2017 in Venice, Italy.

Abstract

Deep globally constrained MRFs for Human Pose Estimation

Ioannis Marras, Petar Palasek, Ioannis Patras

This work introduces a novel Convolutional Network architecture (ConvNet) for the task of human pose estimation, that is the localization of body joints in a single static image. We propose a coarse to fine architecture that addresses shortcomings of the baseline architecture in [Tompson2014] that stem from the fact that large inaccuracies of its coarse ConvNet cannot be corrected by the refinement ConvNet that refines the estimation within small windows of the coarse prediction. We overcome this by introducing a Markov Random Field (MRF)-based spatial model network between the coarse and the refinement model that introduces geometric constraints on the relative locations of the body joints. We propose an architecture in which a) the filters that implement the message passing in the MRF inference are factored in a way that constrains them by a low dimensional pose manifold the projection to which is estimated by a separate branch of the proposed ConvNet and b) the strengths of the pairwise joint constraints are modeled by weights that are jointly estimated by the other parameters of the network. The proposed network is trained in an end-to-end fashion. Experimental results show that the proposed method improves the baseline model and provides state of the art results on very challenging benchmarks.