Angela Yao

Hough Forests for Object Detection, Tracking, and Action Recognition

The paper introduces Hough forests, which are random forests adapted to perform a generalized Hough transform in an efficient way. Compared to previous Hough-based systems such as implicit shape models, Hough forests improve the performance of the generalized Hough transform for object detection on a categorical level. At the same time, their flexibility permits extensions of the Hough transform to new domains such as object tracking and action recognition. Hough forests can be regarded as task-adapted codebooks of local appearance that allow fast supervised training and fast matching at test time. They achieve high detection accuracy since the entries of such codebooks are optimized to cast Hough votes with small variance and since their efficiency permits dense sampling of local image patches or video cuboids during detection. The efficacy of Hough forests for a set of computer vision tasks is validated through experiments on a large set of publicly available benchmark data sets and comparisons with the state-of-the-art.

A Hough transform-based voting framework for action recognition

We present a method to classify and localize human actions in video using a Hough transform voting framework. Random trees are trained to learn a mapping between densely-sampled feature patches and their corresponding votes in a spatio-temporal-action Hough space. The leaves of the trees form a discriminative multi-class codebook that share features between the action classes and vote for action centers in a probabilistic manner. Using low-level features such as gradients and optical flow, we demonstrate that Hough-voting can achieve state-of-the-art performance on several datasets covering a wide range of action-recognition scenarios.

Does human action recognition benefit from pose estimation

Early works on human action recognition focused on tracking and classifying articulated body motions. Such methods required accurate localisation of body parts, which is a difficult task, particularly under realistic imaging conditions. As such, recent trends have shifted towards the use of more abstract, low-level appearance features such as spatio-temporal interest points. Motivated by the recent progress in pose estimation, we feel that pose-based action recognition systems warrant a second look. In this paper, we address the question of whether pose estimation is useful for action recognition or if it is better to train a classifier only on low-level appearance features drawn from video data. We compare pose-based, appearance-based and combined pose and appearance features for action recognition in a home-monitoring scenario. Our experiments show that posebased features outperform low-level appearance features, even when heavily corrupted by noise, suggesting that pose estimation is beneficial for the action recognition task.

Coupled Action Recognition and Pose Estimation from Multiple Views

Action recognition and pose estimation are two closely related topics in understanding human body movements; information from one task can be leveraged to assist the other, yet the two are often treated separately. We present here a framework for coupled action recognition and pose estimation by formulating pose estimation as an optimization over a set of action-specific manifolds. The framework allows for integration of a 2D appearance-based action recognition system as a prior for 3D pose estimation and for refinement of the action labels using relational pose features based on the extracted 3D poses. Our experiments show that our pose estimation system is able to estimate body poses with high degrees of freedom using very few particles and can achieve state-of-the-art results on the HumanEva-II benchmark. We also thoroughly investigate the impact of pose estimation and action recognition accuracy on each other on the challenging TUM kitchen dataset. We demonstrate not only the feasibility of using extracted 3D poses for action recognition, but also improved performance in comparison to action recognition using low-level appearance features.

Variations of a hough-voting action recognition system

This paper presents two variations of a Hough-voting framework used for action recognition and shows classification results for low-resolution video and videos depicting human interactions. For lowresolution videos, where people performing actions are around 30 pixels, we adopt low-level features such as gradients and optical flow. For group actions with human-human interactions, we take the probabilistic action labels from the Hough-voting framework for single individuals and combine them into group actions using decision profiles and classifier combination.

2D action recognition serves 3D human pose estimation

3D human pose estimation in multi-view settings benefits from embeddings of human actions in low-dimensional manifolds, but the complexity of the embeddings increases with the number of actions. Creating separate, action-specific manifolds seems to be a more practical solution. Using multiple manifolds for pose estimation, however, requires a joint optimization over the set of manifolds and the human pose embedded in the manifolds. In order to solve this problem, we propose a particle-based optimization algorithm that can efficiently estimate human pose even in challenging in-house scenarios. In addition, the algorithm can directly integrate the results of a 2D action recognition system as prior distribution for optimization. In our experiments, we demonstrate that the optimization handles an 84D search space and provides already competitive results on HumanEva with as few as 25 particles.

Interactive object detection

In recent years, the rise of digital image and video data available has led to an increasing demand for image annotation. In this paper, we propose an interactive object annotation method that incrementally trains an object detector while the user provides annotations. In the design of the system, we have focused on minimizing human annotation time rather than pure algorithm learning performance. To this end, we optimize the detector based on a realistic annotation cost model based on a user study. Since our system gives live feedback to the user by detecting objects on the fly and predicts the potential annotation costs of unseen images, data can be efficiently annotated by a single user without excessive waiting time. In contrast to popular tracking-based methods for video annotation, our method is suitable for both still images and video. We have evaluated our interactive annotation approach on three datasets, ranging from surveillance, television, to cell microscopy.

Gesture Recognition Portfolios for Personalization

Human gestures, similar to speech and handwriting, are often unique to the individual. Training a generic classifier applicable to everyone can be very difficult and as such, it has become a standard to use personalized classifiers in speech and handwriting recognition. In this paper, we address the problem of personalization in the context of gesture recognition, and propose a novel and extremely efficient way of doing personalization. Unlike conventional personalization methods which learn a single classifier that later gets adapted, our approach learns a set (portfolio) of classifiers during training, one of which is selected for each test subject based on the personalization data. We formulate classifier personalization as a selection problem and propose several algorithms to compute the set of candidate classifiers. Our experiments show that such an approach is much more efficient than adapting the classifier parameters but can still achieve comparable or better results.

Hand Pose Estimation from Local Surface Normals

We present a hierarchical regression framework for estimating hand joint positions from single depth images based on local surface normals. The hierarchical regression follows the tree structured topology of hand from wrist to finger tips. We propose a conditional regression forest, i.e. the Frame Conditioned Regression Forest (FCRF) which uses a new normal difference feature. At each stage of the regression, the frame of reference is established from either the local surface normal or previously estimated hand joints. By making the regression with respect to the local frame, the pose estimation is more robust to rigid transformations. We also introduce a new efficient approximation to estimate surface normals. We verify the effectiveness of our method by conducting experiments on two challenging real-world datasets and show consistent improvements over previous discriminative pose estimation methods.

Tracking people in broadcast sports

We present a method for tracking people in monocular broadcast sports videos by coupling a particle filter with a vote-based confidence map of athletes, appearance features and optical flow for motion estimation. The confidencemap provides a continuous estimate of possible target locations in each frame and outperforms tracking with discrete target detections. We demonstrate the tracker on sports videos, tracking fast and articulated movements of athletes such as divers and gymnasts and on nonsports videos, tracking pedestrians in a PETS2009 sequence.