Pedestrian detection is an essential but challenging task in computer vision, especially in crowded scenes due to heavy intra-class occlusion. In human visual system, head information can be used to locate pedestrian in a crowd because it is more stable and less likely to be occluded. Inspired by this clue, we propose a dual-task detector which detects head and human body simultaneously. Concretely, we estimate human body candidates from head regions with statistical head-body ratio. A head-body alignment map is proposed to perform relational learning between human bodies and heads based on their inherent correlation. We leverage the head information as a strict detection criterion to suppress common false positives of pedestrian detection via a novel pull-push loss. We validate the effectiveness of the proposed method on the CrowdHuman and CityPersons benchmarks. Experimental results demonstrate that the proposed method achieves impressive performance in detecting heavy-occluded pedestrians with little additional computation cost.

The increase in computation cost and storage of convolutional neural networks (CNNs) severely hinders their applications on limited-resources devices in recent years. As a result, there is impending necessity to accelerate the networks by certain methods. In this paper, we propose a loss-driven method to prune redundant channels of CNNs. It identifies unimportant channels by using Taylor expansion technique regarding to scaling and shifting factors, and prunes those channels by fixed percentile threshold. By doing so, we obtain a compact network with less parameters and FLOPs consumption. In experimental section, we evaluate the proposed method in CIFAR datasets with several popular networks, including VGG-19, DenseNet-40 and ResNet-164, and experimental results demonstrate the proposed method is able to prune over 70% channels and parameters with no performance loss. Moreover, iterative pruning could be used to obtain more compact network.

In consideration of the image noise captured by photoelectric cameras at nighttime, a robust motion detection algorithm based on sparse representation is proposed in this study. A universal dictionary for arbitrary scenes is presented. Realistic and synthetic experiments demonstrate the robustness of the proposed approach.

Pedestrian detection is a critical problem in computer vision with significant impact on many real-world applications. In this paper, we introduce an fast dual-task pedestrian detector with integrated segmentation context (DTISC) which predicts pedestrian location as well as its pixel-wise segmentation. The proposed network has three branches where two main branches can independently complete their tasks while useful representations from each task are shared between two branches via the integration branch. Each branch is based on fully convolutional network and is proven effective in its own task. We optimize the detection and segmentation branch on separate ground truths. With reasonable connections, the shared features introduce additional supervision and clues into each branch. Consequently, the two branches are infused at feature spaces increasing their robustness and comprehensiveness. Extensive experiments on pedestrian detection and segmentation benchmarks demonstrate that our joint model improves the performance of detection and segmentation against state-of-the-art algorithms.

A cascaded video denoising method based on frame averaging is proposed in this paper. A novel segmentation approach using intensity and structure tensor is used for change compensation, which can effectively suppress noise while preserving the structure of an image. The cascaded framework solves the problem of noise residual caused by single-frame averaging. The classical Wiener filter is used for spatial denoising in changing areas. Our algorithm works in real-time on an FPGA, since it does not involve future frames. Experiments on standard grayscale videos for various noise levels demonstrate that the proposed method is competitive with current state-of-the-art video denoising algorithms on both peak signal-to-noise ratio and structural similarity evaluations, particularly when dealing with large-scale noise.