A unified framework is proposed to select features by optimizing computationally feasible approximations of high-dimensional conditional mutual information (CMI) between features and their associated class label under different assumptions. Under this unified framework, state-of-the-art information theory based feature selection algorithms are rederived, and a new algorithm is proposed to select features by optimizing a lower bound of the CMI with a weaker assumption than those adopted by existing methods. The new feature selection method integrates a plug-in component to distinguish redundant features from irrelevant ones for improving the feature selection robustness. Furthermore, a novel metric is proposed to evaluate feature selection methods based on simulated data. The proposed method has been compared with state-of-the-art feature selection methods based on the new evaluation metric and classification performance of classifiers built upon the selected features. The experiment results have demonstrated that the proposed method could achieve promising performance in a variety of feature selection problems.

Four-variable-independent-regression localization losses, such as Smooth-1 Loss, are used by default in modern detectors. Nevertheless, this kind of loss is oversimplified so that it is inconsistent with the final evaluation metric, intersection over union (IoU). Directly employing the standard IoU is also not infeasible, since the constant-zero plateau in the case of non-overlapping boxes and the non-zero gradient at the minimum may make it not trainable. Accordingly, we propose a systematic method to address these problems. Firstly, we propose a new metric, the extended IoU (EIoU), which is well-defined when two boxes are not overlapping and reduced to the standard IoU when overlapping. Secondly, we present the convexification technique (CT) to construct a loss on the basis of EIoU, which can guarantee the gradient at the minimum to be zero. Thirdly, we propose a steady optimization technique (SOT) to make the fractional EIoU loss approaching the minimum more steadily and smoothly. Fourthly, to fully exploit the capability of the EIoU based loss, we introduce an interrelated IoU-predicting head to further boost localization accuracy. With the proposed contributions, the new method incorporated into Faster R-CNN with ResNet50+FPN as the backbone yields 4.2 mAP gain on VOC2007 and 2.3 mAP gain on COCO2017 over the baseline Smooth-1 Loss, at almost no training and inferencing computational cost. Specifically, the stricter the metric is, the more notable the gain is, improving 8.2 mAP on VOC2007 and 5.4 mAP on COCO2017 at metric AP90.

Face detection is to search all the possible regions for faces in images and locate the faces if there are any. Many applications including face recognition, facial expression recognition, face tracking and head-pose estimation assume that both the location and the size of faces are known in the image. In recent decades, researchers have created many typical and efficient face detectors from the Viola-Jones face detector to current CNNbased ones. However, with the tremendous increase in images and videos with variations in face scale, appearance, expression, occlusion and pose, traditional face detectors are challenged to detect various "in the wild" faces. The emergence of deep learning techniques brought remarkable breakthroughs to face detection along with the price of a considerable increase in computation. This paper introduces representative deep learning-based methods and presents a deep and thorough analysis in terms of accuracy and efficiency. We further compare and discuss the popular and challenging datasets and their evaluation metrics. A comprehensive comparison of several successful deep learningbased face detectors is conducted to uncover their efficiency using two metrics: FLOPs and latency. The paper can guide to choose appropriate face detectors for different applications and also to develop more efficient and accurate detectors.