Biblio

Video analytics has many applications in traffic control, security monitoring, action/event analysis, etc. With the adoption of deep neural networks, the accuracy of video analytics in video streams has been greatly improved. However, deep neural networks for performing video analytics are compute-intensive. In order to reduce processing time, many systems switch to the lower frame rate or resolution. State-of-the-art switching approaches adjust configurations by profiling video clips on a large configuration space. Multiple configurations are tested periodically and the cheapest one with a desired accuracy is adopted. In this paper, we propose a method that adapts the configuration by analyzing past video analytics results instead of profiling candidate configurations. Our method adopts a lower/higher resolution or frame rate when objects move slow/fast. We train a model that automatically selects the best configuration. We evaluate our method with two real-world video analytics applications: traffic tracking and pose estimation. Compared to the periodic profiling method, our method achieves 3%-12% higher accuracy with the same resource cost and 8-17x faster with comparable accuracy.

Mobile Edge Computing may become a prevalent platform to support applications where mobile devices have limited compute, storage, energy and/or data privacy concerns. In this paper, we study the efficient provisioning and management of compute resources in the Edge-to-Cloud continuum for different types of real-time applications with timeliness requirements depending on application-level update rates and communication/compute delays. We begin by introducing a highly stylized network model allowing us to study the salient features of this problem including its sensitivity to compute vs. communication costs, application requirements, and traffic load variability. We then propose an online decentralized service placement algorithm, based on estimating network delays and adapting application update rates, which achieves high service availability. Our results exhibit how placement can be optimized and how a load-balancing strategy can achieve near-optimal service availability in large networks.

Nearest neighbor search algorithm plays a very important role in computer image algorithm. When the search data is large, we need to use fast search algorithm. The current fast retrieval algorithms are tree based algorithms. The efficiency of the tree algorithm decreases sharply with the increase of the data dimension. In this paper, a local integral hash nearest neighbor algorithm of the spatial space is proposed to construct the tree structure by changing the way of the node of the access tree. It is able to express data distribution characteristics. After experimental testing, this paper achieves more efficient performance in high dimensional data.

This paper focuses on a practically very important problem of matching a real-world product photo to exactly the same item(s) in online shopping sites. The task is extremely challenging because the user photos (i.e., the queries in this scenario) are often captured in uncontrolled environments, while the product images in online shops are mostly taken by professionals with clean backgrounds and perfect lighting conditions. To tackle the problem, we study deep network architectures and training schemes, with the goal of learning a robust deep feature representation that is able to bridge the domain gap between the user photos and the online product images. Our contributions are two-fold. First, we propose an alternative of the popular contrastive loss used in siamese deep networks, namely robust contrastive loss, where we "relax" the penalty on positive pairs to alleviate over-fitting. Second, a multi-task fine-tuning approach is introduced to learn a better feature representation, which not only incorporates knowledge from the provided training photo pairs, but also explores additional information from the large ImageNet dataset to regularize the fine-tuning procedure. Experiments on two challenging real-world datasets demonstrate that both the robust contrastive loss and the multi-task fine-tuning approach are effective, leading to very promising results with a time cost suitable for real-time retrieval.