Danhang Tang

Real-Time Articulated Hand Pose Estimation Using Semi-supervised Transductive Regression Forests

This paper presents the first semi-supervised transductive algorithm for real-time articulated hand pose estimation. Noisy data and occlusions are the major challenges of articulated hand pose estimation. In addition, the discrepancies among realistic and synthetic pose data undermine the performances of existing approaches that use synthetic data extensively in training. We therefore propose the Semi-supervised Transductive Regression (STR) forest which learns the relationship between a small, sparsely labelled realistic dataset and a large synthetic dataset. We also design a novel data-driven, pseudo-kinematic technique to refine noisy or occluded joints. Our contributions include: (i) capturing the benefits of both realistic and synthetic data via transductive learning, (ii) showing accuracies can be improved by considering unlabelled data, and (iii) introducing a pseudo-kinematic technique to refine articulations efficiently. Experimental results show not only the promising performance of our method with respect to noise and occlusions, but also its superiority over state-of-the-arts in accuracy, robustness and speed.

Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture

In this paper we present the Latent Regression Forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards, our method can be considered as a structured coarse-to-fine search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The searching process is guided by a learnt Latent Tree Model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for structured search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180K annotated images from 10 different subjects. Our experiments show that the LRF out-performs state-of-the-art methods in both accuracy and efficiency.

Latent-Class Hough Forests for 3D Object Detection and Pose Estimation

In this paper we propose a novel framework, Latent-Class Hough Forests, for 3D object detection and pose estimation in heavily cluttered and occluded scenes. Firstly, we adapt the state-of-the-art template matching feature, LINEMOD [14], into a scale-invariant patch descriptor and integrate it into a regression forest using a novel template-based split function. In training, rather than explicitly collecting representative negative samples, our method is trained on positive samples only and we treat the class distributions at the leaf nodes as latent variables. During the inference process we iteratively update these distributions, providing accurate estimation of background clutter and foreground occlusions and thus a better detection rate. Furthermore, as a by-product, the latent class distributions can provide accurate occlusion aware segmentation masks, even in the multi-instance scenario. In addition to an existing public dataset, which contains only single-instance sequences with large amounts of clutter, we have collected a new, more challenging, dataset for multiple-instance detection containing heavy 2D and 3D clutter as well as foreground occlusions. We evaluate the Latent-Class Hough Forest on both of these datasets where we outperform state-of-the art methods.

Fast Pedestrian Detection by Cascaded Random Forest with Dominant Orientation Templates.

In this paper, we present a new pedestrian detection method combining Random Forest and Dominant Orientation Templates(DOT) to achieve state-of-the-art accuracy and, more importantly, to accelerate run-time speed. DOT can be considered as a binary version of Histogram of Oriented Gradients(HOG) and therefore provides time-efficient properties. However, since discarding magnitude information, it degrades the detection rate, when it is directly incorporated. We propose a novel template-matching split function using DOT for Random Forest. It divides a feature space in a non-linear manner, but has a very low complexity up to binary bit-wise operations. Experiments demonstrate that our method provides much superior speed with comparable accuracy to state-ofthe-art pedestrian detectors. By combining a holistic and a patch-based detectors in a cascade manner, we accelerate the detection speed of Hough Forest, a prior-art using Random Forest and HOG, by about 20 times. The obtained speed is 5 frames per second for 640×480 images with 24 scales.

Conditional Convolutional Neural Network for Modality-Aware Face Recognition

Faces in the wild are usually captured with various poses, illuminations and occlusions, and thus inherently multimodally distributed in many tasks. We propose a conditional Convolutional Neural Network, named as c-CNN, to handle multimodal face recognition. Different from traditional CNN that adopts fixed convolution kernels, samples in c-CNN are processed with dynamically activated sets of kernels. In particular, convolution kernels within each layer are only sparsely activated when a sample is passed through the network. For a given sample, the activations of convolution kernels in a certain layer are conditioned on its present intermediate representation and the activation status in the lower layers. The activated kernels across layers define the sample-specific adaptive routes that reveal the distribution of underlying modalities. Consequently, the proposed framework does not rely on any prior knowledge of modalities in contrast with most existing methods. To substantiate the generic framework, we introduce a special case of c-CNN via incorporating the conditional routing of the decision tree, which is evaluated with two problems of multimodality - multi-view face identification and occluded face verification. Extensive experiments demonstrate consistent improvements over the counterparts unaware of modalities.

Spatio-Temporal Hough Forest for efficient detection–localisation–recognition of fingerwriting in egocentric camera

Abstract Recognising fingerwriting in mid-air is a useful input tool for wearable egocentric camera. In this paper we propose a novel framework to this purpose. Specifically, our method first detects a writing hand posture and locates the position of index fingertip in each frame. From the trajectory of the fingertip, the written character is localised and recognised simultaneously. To achieve this challenging task, we first present a contour-based view independent hand posture descriptor extracted with a novel signature function. The proposed descriptor serves both posture recognition and fingertip detection. As to recognising characters from trajectories, we propose Spatio-Temporal Hough Forest that takes sequential data as input and perform regression on both spatial and temporal domain. Therefore our method can perform character recognition and localisation simultaneously. To establish our contributions, a new handwriting-in-mid-air dataset with labels for postures, fingertips and character locations is proposed. We design and conduct experiments of posture estimation, fingertip detection, character recognition and localisation. In all experiments our method demonstrates superior accuracy and robustness compared to prior arts.

Latent Regression Forest: Structured Estimation of 3D Hand Poses

In this paper we present the latent regression forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. Prior discriminative methods often fall into two categories: holistic and patch-based. Holistic methods are efficient but less flexible due to their nearest neighbour nature. Patch-based methods can generalise to unseen samples by consider local appearance only. However, they are complex because each pixel need to be classified or regressed during testing. In contrast to these two baselines, our method can be considered as a structured coarse-to-fine search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The searching process is guided by a learnt latent tree model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for structured search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180 K annotated images from 10 different subjects. Our experiments on two datasets show that the LRF outperforms baselines and prior arts in both accuracy and efficiency.

Latent-Class Hough Forests for 6 DoF Object Pose Estimation

In this paper we present Latent-Class Hough Forests, a method for object detection and 6 DoF pose estimation in heavily cluttered and occluded scenarios. We adapt a state of the art template matching feature into a scale-invariant patch descriptor and integrate it into a regression forest using a novel template-based split function. We train with positive samples only and we treat class distributions at the leaf nodes as latent variables. During testing we infer by iteratively updating these distributions, providing accurate estimation of background clutter and foreground occlusions and, thus, better detection rate. Furthermore, as a by-product, our Latent-Class Hough Forests can provide accurate occlusion aware segmentation masks, even in the multi-instance scenario. In addition to an existing public dataset, which contains only single-instance sequences with large amounts of clutter, we have collected two, more challenging, datasets for multiple-instance detection containing heavy 2D and 3D clutter as well as foreground occlusions. We provide extensive experiments on the various parameters of the framework such as patch size, number of trees and number of iterations to infer class distributions at test time. We also evaluate the Latent-Class Hough Forests on all datasets where we outperform state of the art methods.

A comparative study of video-based object recognition from an egocentric viewpoint

Videos tend to yield a more complete description of their content than individual images. And egocentric vision often provides a more controllable and practical perspective for capturing useful information. In this study, we presented new insights into different object recognition methods for video-based rigid object instance recognition. In order to better exploit egocentric videos as training and query sources, diverse state-of-the-art techniques were categorised, extended and evaluated empirically using a newly collected video dataset, which consists of complex sculptures in clutter scenes. In particular, we investigated how to utilise the geometric and temporal cues provided by egocentric video sequences to improve the performance of object recognition. Based on the experimental results, we analysed the pros and cons of these methods and reached the following conclusions. For geometric cues, the 3D object structure learnt from a training video dataset improves the average video classification performance dramatically. By contrast, for temporal cues, tracking visual fixation among video sequences has little impact on the accuracy, but significantly reduces the memory consumption by obtaining a better signal-to-noise ratio for the feature points detected in the query frames. Furthermore, we proposed a method that integrated these two important cues to exploit the advantages of both.

Opening the Black Box: Hierarchical Sampling Optimization for Hand Pose Estimation

Hand pose estimation, formulated as an inverse problem, is typically optimized by an energy function over pose parameters using a ‘black box’ image generation procedure, knowing little about either the relationships between the parameters or the form of the energy function. In this paper, we show significant improvement upon such black box optimization by exploiting high-level knowledge of the parameter structure and using a local surrogate energy function. Our new framework, hierarchical sampling optimization (HSO), consists of a sequence of discriminative predictors organized into a kinematic hierarchy. Each predictor is conditioned on its ancestors, and generates a set of samples over a subset of the pose parameters, with only one selected by the highly-efficient surrogate energy. The selected partial poses are concatenated to generate a full-pose hypothesis. Repeating the same process, several hypotheses are generated and the full energy function selects the best result. Under the same kinematic hierarchy, two methods based on decision forest and convolutional neural network are proposed to generate the samples and two optimization methods are studied when optimizing these samples. Experimental evaluations on three publicly available datasets show that our method is particularly impressive in low-compute scenarios where it significantly outperforms all other state-of-the-art methods.