Integrating multiple different yet complementary feature representations has been proved to be an effective way for boosting tracking performance. This paper investigates how to perform robust object tracking in challenging scenarios by adaptively incorporating information from grayscale and thermal videos, and proposes a novel collaborative algorithm for online tracking. In particular, an adaptive fusion scheme is proposed based on collaborative sparse representation in Bayesian filtering framework. We jointly optimize sparse codes and the reliable weights of different modalities in an online way. In addition, this paper contributes a comprehensive video benchmark, which includes 50 grayscale-thermal sequences and their ground truth annotations for tracking purpose. The videos are with high diversity and the annotations were finished by one single person to guarantee consistency. Extensive experiments against other state-of-the-art trackers with both grayscale and grayscale-thermal inputs demonstrate the effectiveness of the proposed tracking approach. Through analyzing quantitative results, we also provide basic insights and potential future research directions in grayscale-thermal tracking.

No abstract

Abstract-Elastic partitioning of computations between mobile devices and cloud is an important and challenging research topic for mobile cloud computing. Existing works focus on the single-user computation partitioning, which aims to optimize the application completion time for one particular single user. These works assume that the cloud always has enough resources to execute the computations immediately when they are offloaded to the cloud. However, this assumption does not hold for large scale mobile cloud applications. In these applications, due to the competition for cloud resources among a large number of users, the offloaded computations may be executed with certain scheduling delay on the cloud. Single user partitioning that does not take into account the scheduling delay on the cloud may yield significant performance degradation. In this paper, we study, for the first time, Multi-user Computation Partitioning Problem (MCPP), which considers the partitioning of multiple users' computations together with the scheduling of offloaded computations on the cloud resources. Instead of pursuing the minimum application completion time for every single user, we aim to achieve minimum average completion time for all the users, based on the number of provisioned resources on the cloud. We show that MCPP is different from and more difficult than the classical job scheduling problems. We design an offline heuristic algorithm, namely SearchAdjust, to solve MCPP. We demonstrate through benchmarks that SearchAdjust outperforms both the single user partitioning approaches and classical job scheduling approaches by 10% on average in terms of application delay. Based on SearchAdjust, we also design an online algorithm for MCPP that can be easily deployed in practical systems. We validate the effectiveness of our online algorithm using real world load traces.

Semantic labeling of RGB-D scenes is crucial to many intelligent applications including perceptual robotics. It generates pixelwise and fine-grained label maps from simultaneously sensed photometric (RGB) and depth channels. This paper addresses this problem by i) developing a novel Long Short-Term Memorized Context Fusion (LSTM-CF) Model that captures and fuses contextual information from multiple channels of photometric and depth data, and ii) incorporating this model into deep convolutional neural networks (CNNs) for end-to-end training. Specifically, contexts in photometric and depth channels are, respectively, captured by stacking several convolutional layers and a long short-term memory layer; the memory layer encodes both short-range and longrange spatial dependencies in an image along the vertical direction. Another long short-term memorized fusion layer is set up to integrate the contexts along the vertical direction from different channels, and perform bi-directional propagation of the fused vertical contexts along the horizontal direction to obtain true 2D global contexts. At last, the fused contextual representation is concatenated with the convolutional features extracted from the photometric channels in order to improve the accuracy of fine-scale semantic labeling. Our proposed model has set a new state of the art, i.e., 48.1% and 49.4% average class accuracy over 37 categories (2.2% and 5.4% improvement) on the large-scale SUNRGBD dataset and the NYUDv2 dataset, respectively.

No abstract

No abstract

We propose to use action, scene and object concepts as semantic attributes for classification of video events in InTheWild content, such as YouTube videos. We model events using a variety of complementary semantic attribute features developed in a semantic concept space. Our contribution is to systematically demonstrate the advantages of this concept-based event representation (CBER) in applications of video event classification and understanding. Specifically, CBER has better generalization capability, which enables to recognize events with a few training examples. In addition, CBER makes it possible to recognize a novel event without training examples (i.e., zero-shot learning). We further show our proposed enhanced event model can further improve the zero-shot learning. Furthermore, CBER provides a straightforward way for event recounting/understanding. We use the TRECVID Multimedia Event Detection (MED11) open source event definitions and datasets as our test bed and show results on over 1400 hours of videos.

3D human articulated pose recovery from monocular image sequences is very challenging due to the diverse appearances, viewpoints, occlusions, and also the human 3D pose is inherently ambiguous from the monocular imagery. It is thus critical to exploit rich spatial and temporal long-range dependencies among body joints for accurate 3D pose sequence prediction. Existing approaches usually manually design some elaborate prior terms and human body kinematic constraints for capturing structures, which are often insufficient to exploit all intrinsic structures and not scalable for all scenarios. In contrast, this paper presents a Recurrent 3D Pose Sequence Machine(RPSM) to automatically learn the image-dependent structural constraint and sequence-dependent temporal context by using a multi-stage sequential refinement. At each stage, our RPSM is composed of three modules to predict the 3D pose sequences based on the previously learned 2D pose representations and 3D poses: (i) a 2D pose module extracting the image-dependent pose representations, (ii) a 3D pose recurrent module regressing 3D poses and (iii) a feature adaption module serving as a bridge between module (i) and (ii) to enable the representation transformation from 2D to 3D domain. These three modules are then assembled into a sequential prediction framework to refine the predicted poses with multiple recurrent stages. Extensive evaluations on the Human3.6M dataset and HumanEva-I dataset show that our RPSM outperforms all state-of-the-art approaches for 3D pose estimation. * Corresponding author is Liang Lin.