Video rain/snow removal from surveillance videos is an important task in the computer vision community since rain/snow existed in videos can severely degenerate the performance of many surveillance system. Various methods have been investigated extensively, but most only consider consistent rain/snow under stable background scenes. Rain/snow captured from practical surveillance camera, however, is always highly dynamic in time with the background scene transformed occasionally. To this issue, this paper proposes a novel rain/snow removal approach, which fully considers dynamic statistics of both rain/snow and background scenes taken from a video sequence. Specifically, the rain/snow is encoded as an online multi-scale convolutional sparse coding (OMS-CSC) model, which not only finely delivers the sparse scattering and multi-scale shapes of real rain/snow, but also well encodes their temporally dynamic configurations by real-time ameliorated parameters in the model. Furthermore, a transformation operator imposed on the background scenes is further embedded into the proposed model, which finely conveys the dynamic background transformations, such as rotations, scalings and distortions, inevitably existed in a real video sequence. The approach so constructed can naturally better adapt to the dynamic rain/snow as well as background changes, and also suitable to deal with the streaming video attributed its online learning mode. The proposed model is formulated in a concise maximum a posterior (MAP) framework and is readily solved by the ADMM algorithm. Compared with the state-of-the-art online and offline video rain/snow removal methods, the proposed method achieves better performance on synthetic and real videos datasets both visually and quantitatively. Specifically, our method can be implemented in relatively high efficiency, showing its potential to real-time video rain/snow removal.Index Terms-multi-scale, convolutional sparse coding, rain/snow removal, online learning, alignment method.

Randomness is critical for many information processing applications, including numerical modeling and cryptography [1,2]. Device-independent quantum random number generation [3] (DIQRNG) based on the loophole free violation of Bell inequality [4][5][6][7] produces unpredictable genuine randomness without any device assumption and is therefore an ultimate goal in the field of quantum information science [8][9][10]. However, due to formidable technical challenges, there were very few reported experimental studies of DIQRNG [11][12][13][14], which were vulnerable to the adversaries. Here we present a fully functional DIQRNG against the most general quantum adversaries [15][16][17]. We construct a robust experimental platform that realizes Bell inequality violation with entangled photons with detection and locality loopholes closed simultaneously. This platform enables a continuous recording of a large volume of data sufficient for security analysis against the general quantum side information and without assuming independent and identical distribution.Lastly, by developing a large Toeplitz matrix (137.90 Gb × 62.469 Mb) hashing technique, we demonstrate that this DIQRNG generates 6.2469 × 10 7 quantum-certified random bits in 96 hours (or 181 bits/s) with uniformity within 10 −5 . We anticipate this DIQRNG may have profound impact on the research of quantum randomness and information-secured applications.

This is a repository copy of Device-independent randomness expansion against quantum side information.

No abstract