Abstract: Recognizing human actions in videos is a challenging problem in computer vision. Recently, convolutional neural network based deep features have shown promising results for action recognition. In this paper, we investigate the problem of fusing deep appearance and motion cues for action recognition. We propose a video representation which combines deep appearance and motion based local convolutional features within the bag-of-deep-features framework. Firstly, dense deep appearance and motion based local convolutional features are extracted from spatial (RGB) and temporal (flow) networks, respectively. Both visual cues are processed in parallel by constructing separate visual vocabularies for appearance and motion. A category-specific appearance map is then learned to modulate the weights of the deep motion features. The proposed representation is discriminative and binds the deep local convolutional features to their spatial locations. Experiments are performed on two challenging datasets: JHMDB dataset with 21 action classes and ACT dataset with 43 categories. The results clearly demonstrate that our approach outperforms both standard approaches of early and late feature fusion. Further, our approach is only employing action labels and without exploiting body part information, but achieves competitive performance compared to the state-of-the-art deep features based approaches.

Abstract: Depth can complement RGB with useful cues about object volumes and scene layout. However, RGB-D image datasets are still too small for directly training deep convolutional neural networks (CNNs), in contrast to the massive monomodal RGB datasets. Previous works in RGB-D recognition typically combine two separate networks for RGB and depth data, pretrained with a large RGB dataset and then fine tuned to the respective target RGB and depth datasets. These approaches have several limitations: 1) only use low-level filters learned from RGB data, thus not being able to exploit properly depth-specific patterns, and 2) RGB and depth features are only combined at high-levels but rarely at lower-levels. In this paper, we propose a framework that leverages both knowledge acquired from large RGB datasets together with depth-specific cues learned from the limited depth data, obtaining more effective multi-source and multi-modal representations. We propose a multi-modal combination method that selects discriminative combinations of layers from the different source models and target modalities, capturing both high-level properties of the task and intrinsic low-level properties of both modalities.

Abstract: Pooling is a ubiquitous operation in image processing algorithms that allows for higher-level processes to collect relevant low-level features from a region of interest. Currently, max-pooling is one of the most commonly used operators in the computational literature. However, it can lack robustness to outliers due to the fact that it relies merely on the peak of a function. Pooling mechanisms are also present in the primate visual cortex where neurons of higher cortical areas pool signals from lower ones. The receptive fields of these neurons have been shown to vary according to the contrast by aggregating signals over a larger region in the presence of low contrast stimuli. We hypothesise that this contrast-variant-pooling mechanism can address some of the shortcomings of maxpooling. We modelled this contrast variation through a histogram clipping in which the percentage of pooled signal is inversely proportional to the local contrast of an image. We tested our hypothesis by applying it to the phenomenon of colour constancy where a number of popular algorithms utilise a max-pooling step (e.g. White-Patch, Grey-Edge and Double-Opponency). For each of these methods, we investigated the consequences of replacing their original max-pooling by the proposed contrast-variant-pooling. Our experiments on three colour constancy benchmark datasets suggest that previous results can significantly improve by adopting a contrast-variant-pooling mechanism.

Abstract: Deep learning for human action recognition in videos is making significant progress, but is slowed down by its dependency on expensive manual labeling of large video collections. In this work, we investigate the generation of synthetic training data for action recognition, as it has recently shown promising results for a variety of other computer vision tasks. We propose an interpretable parametric generative model of human action videos that relies on procedural generation and other computer graphics techniques of modern game engines. We generate a diverse, realistic, and physically plausible dataset of human action videos, called PHAV for ”Procedural Human Action Videos”. It contains a total of 39, 982 videos, with more than 1, 000 examples for each action of 35 categories. Our approach is not limited to existing motion capture sequences, and we procedurally define 14 synthetic actions. We introduce a deep multi-task representation learning architecture to mix synthetic and real videos, even if the action categories differ. Our experiments on the UCF101 and HMDB51 benchmarks suggest that combining our large set of synthetic videos with small real-world datasets can boost recognition performance, significantly
outperforming fine-tuning state-of-the-art unsupervised generative models of videos.