Search

In outdoor crimes such as robbery and kidnapping, suspects generally secretly follow their victims in public places and then look for opportunities to commit crimes. Video anomaly detection (VAD) has achieved fruitful results through deep neural networks (DNN). However, as an abnormal behavior without obvious abnormal physical features, hidden following is highly similar to ordinary walking and accompanying behaviors, so it is difficult to effectively detect hidden dangerous followers using video anomaly detection methods or traditional trajectory analysis methods. We propose "hidden follower'' detection (HFD) task and a HFD model based on gaze pattern extraction. It extracts gaze pattern features of pedestrians from gaze-interval-series and introduces a time series classification model to classify pedestrians with or without hidden following purposes. Based on this model, we propose a hidden follower detection framework (HFDF) to detect hidden followers from normal pedestrians, which utilizes the trajectories and gaze patterns extracted from videos. To cope with the lack of test data, we construct a dataset of 1200 pedestrians from the crowd simulation model to simulate scenes including hidden followers, and we also collected a surveillance video dataset including the hidden following behaviors. The experiments conducted on these two datasets show that HFDF can consistently outperform the state-of-the-art method by a notable margin in the HFD task on the commonly-used F1 benchmark.

Vehicle re-identification (re-ID) has received more attention in recent years as a significant work, making huge contribution to the intelligent video surveillance. The complex intra-class and inter-class variation of vehicle images bring huge challenges for vehicle re-ID, especially for the similar vehicle re-ID. In this paper we focus on an interesting and challenging problem, vehicle re-ID of the same/similar model. Previous works mainly focus on extracting global features using deep models, ignoring the individual loa-cal regions in vehicle front window, such as decorations and stickers attached to the windshield, that can be more discriminative for vehicle re-ID. Instead of directly embedding these regions to learn their features, we propose a Regional Structure-Aware model (RSA) to learn structure-aware cues with the position distribution of individual local regions in vehicle front window area, constructing a FW structural map space. In this map sapce, deep models are able to learn more robust and discriminative spatial structure-aware features to improve the performance for vehicle re-ID of the same/similar model. We evaluate our method on a large-scale vehicle re-ID dataset Vehicle-1M. The experimental results show that our method can achieve promising performance and outperforms several recent state-of-the-art approaches.

Most video surveillance suspect investigation systems rely on the videos taken in different camera views. Actually, besides the videos, in the investigation process, investigators also manually label some marks, which, albeit incomplete, can be quite accurate and helpful in identifying persons. This paper studies the problem of Person Re-identification with Incomplete Marks (PRIM), aiming at ranking the persons in the gallery according to both the videos and incomplete marks. This problem is solved by a multi-step fusion algorithm, which consists of three key steps: (i) The early fusing step exploits both visual features and marked attributes to predict a complete and precise attribute vector. (ii) Based on the statistical attribute d ominance and saliency phenomena, a dominance-saliency matching model is suggested for measuring the distance between attribute vectors. (iii) The gallery is ranked separately by using visual features and attribute vectors, and the overall ranking list is the result of a late fusion. Experiments conducted on VIPeR dataset have validated the effectiveness of the proposed method in all the three key steps. The results also show that through introducing marks, the retrieval accuracy is significantly improved.

As a core technique supporting the multi-camera tracking task, person re-identification attracts increasing research interests in both academic and industrial communities. Its aim is to match individuals across a group of spatially non-overlapping surveillance cameras, which are usually interfered by various imaging conditions and object motions. Current methods mainly focus on robust feature representation and accurate distance measure, where intensive computations and expensive training samples prohibit their practical applications. To address the above problems, this paper proposes a new unsupervised person re-identification method featured by its competitive accuracy and high efficiency. Both merits stem from model based person image representation and ranking, with which, merely 4-dimension pixel-level features can achieve over 20% matching rate at Rank 1 on the challenging VIPeR dataset.

Vehicle re-identification, retrieving a vehicle detected by one camera with the same vehicle by another camera, is an important problem in the video investigation application which is a technology for criminal investigation. In this task, it not only needs to classify the vehicle category, but also to identify a specific object in the category. Previous methods mainly focus on the vehicle categorization, which cannot identify the specific vehicle. In this paper, a two-stage strategy is proposed to accomplish vehicle re-identification in realistic surveillance videos. Specifically, in the first stage, a part-based appearance model fusing multiple visual features is proposed to represent the vehicle object, and then a coarse ranking list is generated by comparing appearance models of the probe and gallery vehicles. In the second stage, the temporal-spatial relation is introduced to re-rank the above visual-based ranking list, where vehicles of the same category and reasonable spatial-temporal relations are placed in top positions while those of mismatched types or relations are placed in rear positions. Both quantitative and qualitative experiments conducted on a real world dataset have validated the effectiveness of the proposed method.

In this paper, we propose a manifold learning based algorithm using 'Nearest Feature Line - NFL' to hallucinate high-resolution face image. According to the fact that existing NFL can effectively characterize the geometrical proportions to the face samples, we propose using NFL metric to define the neighborhood relations between face samples. Our algorithm can solve the problem that traditional method cannot effectively reveal the similar local geometry between high-resolution and low-resolution face manifolds under the condition that the training sample size is small. Moreover, in order to enhance the representation capacity of available face samples and reduce the computational complexity, we select neighborhood samples for each input LR image. Experimental results demonstrate that our algorithm can generates clearer local feature details, and the PSNR is 1.4 dB higher than that of the best manifold learning based method reported so far.

In real surveillance scenarios, a variety of factors have an impact on the quality of images, which leads to pixel distortion and aliasing. Traditional face super-resolution algorithms only use the difference of image pixel values as similarity criterion, which degrades similarity and identification of reconstructed facial images. Image semantic information with human understanding, especially structural data of shapes, is robust to the degraded images. In this paper, we propose a face hallucination with shape parameters projection constraint. This method uses a parameter model to represent face shapes, and shape information of input image is introduced to improving the quality of reconstructed image. The shape model regularization is first added to original objective function. Then shape parameters are projected into the domain of image parameters by a linear regression model. Finally, the gradient descent method is used to obtain the unified parameters. Experimental results demonstrate the proposed method outperforms the traditional schemes significantly both in subjective and objective quality.

Existing knowledge on how people use speech-based technologies in realistic settings is limited. We are conducting a longitudinal field study, spanning six months, to investigate how users with no physical impairments and users with upper body physical impairments use speech technologies when interacting with computers in their home environment. Digital data logs, time diaries, and interviews are being used to record the types of applications used, frequency of use of each application, and difficulties experienced as well as subjective data regarding the usage experience. While confirming many expectations, initial results have provided several unexpected insights including a preference to use speech for navigation instead of dictation tasks, and the use of speech technology for programming and games.