arXiv subjects

Cheng Chi

Publications and source records attributed to Cheng Chi.

93 records · Page 6Linked to original sources

arXiv2019

WIDER Face and Pedestrian Challenge 2018: Methods and Results

This paper presents a review of the 2018 WIDER Challenge on Face and Pedestrian. The challenge focuses on the problem of precise localization of human faces and bodies, and accurate association of identities. It comprises of three tracks: (i) WIDER Face which aims at soliciting new approaches to advance the state-of-the-art in face detection, (ii) WIDER Pedestrian which aims to find effective and efficient approaches to address the problem of pedestrian detection in unconstrained environments, and (iii) WIDER Person Search which presents an exciting challenge of searching persons across 192 movies. In total, 73 teams made valid submissions to the challenge tracks. We summarize the winning solutions for all three tracks. and present discussions on open problems and potential research directions in these topics.

cs.CV↗

arXiv2018

Selective Refinement Network for High Performance Face Detection

High performance face detection remains a very challenging problem, especially when there exists many tiny faces. This paper presents a novel single-shot face detector, named Selective Refinement Network (SRN), which introduces novel two-step classification and regression operations selectively into an anchor-based face detector to reduce false positives and improve location accuracy simultaneously. In particular, the SRN consists of two modules: the Selective Two-step Classification (STC) module and the Selective Two-step Regression (STR) module. The STC aims to filter out most simple negative anchors from low level detection layers to reduce the search space for the subsequent classifier, while the STR is designed to coarsely adjust the locations and sizes of anchors from high level detection layers to provide better initialization for the subsequent regressor. Moreover, we design a Receptive Field Enhancement (RFE) block to provide more diverse receptive field, which helps to better capture faces in some extreme poses. As a consequence, the proposed SRN detector achieves state-of-the-art performance on all the widely used face detection benchmarks, including AFW, PASCAL face, FDDB, and WIDER FACE datasets. Codes will be released to facilitate further studies on the face detection problem.

cs.CV↗

arXiv2017

Concurrent intersection point of magnetization and magneto-conductivity curves in strongly fluctuating superconductors

The thermal fluctuations contribution to magnetization and magneto-conductivity of type II layered superconductor is calculated in the framework of Lawrence-Doniach model. For numerous high temperature cuprate superconductors, it was discovered that the magnetization dependence on temperature in wide range of fields exhibits an intersection point at a temperature slightly below $T_{c}$. We notice a similar intersection point of the magneto-conductivity curves at the approximate same temperature. The phenomenon is explained by strong (non-gaussian) thermal fluctuations with interactions treated using a self-consistent theory. All higher Landau levels should be included. Dimensionality of the fluctuations is defined and the 2D-3D dimensional crossover is the key for the existence of intersection points.

cond-mat.supr-con↗