The aim of this paper is to address recognition of natural human actions in diverse and realistic video settings. This challenging but important subject has mostly been ignored in the past due to several problems one of which is the lack of realistic and annotated video datasets. Our first contribution is to address this limitation and to investigate the use of movie scripts for automatic annotation of human actions in videos. We evaluate alternative methods for action retrieval from scripts and show benefits of a text-based classifier. Using the retrieved action samples for visual learning, we next turn to the problem of action classification in video. We present a new method for video classification that builds upon and extends several recent ideas including local space-time features, space-time pyramids and multichannel non-linear SVMs. The method is shown to improve state-of-the-art results on the standard KTH action dataset by achieving 91.8 % accuracy. Given the inherent problem of noisy labels in automatic annotation, we particularly investigate and show high tolerance of our method to annotation errors in the training set. We finally apply the method to learning and classifying challenging action classes in movies and show promising results. 1.

In this work, we present a novel local descriptor for video sequences. The proposed descriptor is based on histograms of oriented 3D spatio-temporal gradients. Our contribution is four-fold. (i) To compute 3D gradients for arbitrary scales, we develop a memory-efficient algorithm based on integral videos. (ii) We propose a generic 3D orientation quantization which is based on regular polyhedrons. (iii) We perform an in-depth evaluation of all descriptor parameters and optimize them for action recognition. (iv) We apply our descriptor to various action datasets (KTH, Weizmann, Hollywood) and show that we outperform the state-of-the-art. 1

Local space-time features have recently become a popular video representation for action recognition. Several methods for feature localization and description have been proposed in the literature and promising recognition results were demonstrated for a number of action classes. The comparison of existing methods, however, is often limited given the different experimental settings used. The purpose of this paper is to evaluate and compare previously proposed space-time features in a common experimental setup. In particular, we consider four different feature detectors and six local feature descriptors and use a standard bag-of-features SVM approach for action recognition. We investigate the performance of these methods on a total of 25 action classes distributed over three datasets with varying difficulty. Among interesting conclusions, we demonstrate that regular sampling of space-time features consistently outperforms all tested space-time interest point detectors for human actions in realistic settings. We also demonstrate a consistent ranking for the majority of methods over different datasets and discuss their advantages and limitations. 1

Abstract. The use of sparse invariant features to recognise classes of actions or objects has become common in the literature. However, features are often ”engineered ” to be both sparse and invariant to transformation and it is assumed that they provide the greatest discriminative information. To tackle activity recognition, we propose learning compound features that are assembled from simple 2D corners in both space and time. Each corner is encoded in relation to its neighbours and from an over complete set (in excess of 1 million possible features), compound features are extracted using data mining. The final classifier, consisting of sets of compound features, can then be applied to recognise and localise an activity in real-time while providing superior performance to other state-of-the-art approaches (including those based upon sparse feature detectors). Furthermore, the approach requires only weak supervision in the form of class labels for each training sequence. No ground truth position or temporal alignment is required during training. 1

In this paper we revisit local feature detectors/descriptors developed for 2D images and extend them to the more general framework of scalar fields defined on 2D manifolds. We provide methods and tools to detect and describe features on surfaces equiped with scalar functions, such as photometric information. This is motivated by the growing need for matching and tracking photometric surfaces over temporal sequences, due to recent advancements in multiple camera 3D reconstruction. We propose a 3D feature detector (MeshDOG) and a 3D feature descriptor (MeshHOG) for uniformly triangulated meshes, invariant to changes in rotation, translation, and scale. The descriptor is able to capture the local geometric and/or photometric properties in a succinct fashion. Moreover, the method is defined generically for any scalar function, e.g., local curvature. Results with matching rigid and non-rigid meshes demonstrate the interest of the proposed framework.

Within the field of action recognition, features and descriptors are often engineered to be sparse and invariant to transformation. While sparsity makes the problem tractable, it is not necessarily optimal in terms of class separability and classification. This paper proposes a novel approach that uses very dense corner features that are spatially and temporally grouped in a hierarchical process to produce an overcomplete compound feature set. Frequently reoccurring patterns of features are then found through data mining, designed for use with large data sets. The novel use of the hierarchical classifier allows real time operation while the approach is demonstrated to handle camera motion, scale, human appearance variations, occlusions and background clutter. The performance of classification, outperforms other state-of-the-art action recognition algorithms on the three datasets; KTH, multi-KTH, and realworld movie sequences containing broad actions. Multiple action localisation is performed, though no groundtruth localisation data is required, using only weak supervision of class labels for each training sequence. The realworld movie dataset contain complex realistic actions from movies, the approach outperforms the published accuracy on this dataset and also achieves real time performance. 1.

Several spatiotemporal feature point detectors have been recently used in video analysis for action recognition. Feature points are detected using a number of measures, namely saliency, cornerness, periodicity, motion activity etc. Each of these measures is usually intensity-based and provides a different trade-off between density and informativeness. In this paper, we use saliency for feature point detection in videos and incorporate color and motion apart from intensity. Our method uses a multi-scale volumetric representation of the video and involves spatiotemporal operations at the voxel level. Saliency is computed by a global minimization process constrained by pure volumetric constraints, each of them being related to an informative visual aspect, namely spatial proximity, scale and feature similarity (intensity, color, motion). Points are selected as the extrema of the saliency response and prove to balance well between density and informativeness. We provide an intuitive view of the detected points and visual comparisons against state-of-the-art space-time detectors. Our detector outperforms them on the KTH dataset using Nearest-Neighbor classifiers and ranks among the top using different classification frameworks. Statistics and comparisons are also performed on the more difficult Hollywood Human Actions (HOHA) dataset increasing the performance compared to current published results. 1.

This paper discusses an approach to human action recognition via local feature tracking and robust estimation of background motion. The main contribution is a robust feature extraction algorithm based on KLT tracker and SIFT as well as a method for estimating dominant planes in the scene. Multiple interest point detectors are used to provide large number of features for every frame. The motion vectors for the features are estimated using optical flow and SIFT based matching. The features are combined with image segmentation to estimate dominant homographies, and then separated into static and moving ones regardless the camera motion. The action recognition approach can handle camera motion, zoom, human appearance variations, background clutter and occlusion. The motion compensation shows very good accuracy on a number of test sequences. The recognition system is extensively compared to state-of-the art action recognition methods and the results are improved. 1

Video understanding is a computationally challenging task that is critical not only for traditionally throughput-oriented applications such as search but also latency-sensitive interactive applications such as surveillance, gaming, videoconferencing, and vision-based user interfaces. Enabling these types of video processing applications will require not only new algorithms and techniques, but new runtime systems that optimize latency as well as throughput. In this paper, we present a runtime system called Sprout that achieves low latency by exploiting the parallelism inherent in video understanding applications. We demonstrate the utility of our system on an activity recognition application that employs a robust new descriptor called MoSIFT, which explicitly augments appearance features with motion information. MoSIFT outperforms previous recognition techniques, but like other state-of-the-art techniques, it is computationally expensive — a sequential implementation runs 100 times slower than real time. We describe the implementation of the activity recognition application on Sprout, and show that it can accurately recognize activities at full frame rate (25 fps) and low latency on a challenging airport surveillance video corpus.

In this paper we address the problem of localisation and recognition of human activities in unsegmented image sequences. The main contribution of the proposed method is the use of an implicit representation of the spatiotemporal shape of the activity which relies on the spatiotemporal localization of characteristic, sparse, ’visual words ’ and ’visual verbs’. Evidence for the spatiotemporal localization of the activity are accumulated in a probabilistic spatiotemporal voting scheme. The local nature of our voting framework allows us to recover multiple activities that take place in the same scene, as well as activities in the presence of clutter and occlusions. We construct class-specific codebooks using the descriptors in the training set, where we take the spatial co-occurrences of pairs of codewords into account. The positions of the codeword pairs with respect to the object centre, as well as the frame in the training set in which they occur are subsequently stored in order to create a spatiotemporal model of codeword co-occurrences. During the testing phase, we use Mean Shift Mode estimation in order to spatially segment the subject that performs the activities in every frame, and the Radon transform in order to extract the most probable hypotheses concerning the temporal segmentation of the activities within the continuous stream. 1.