No abstract

Modern crowd counting methods usually employ deep neural networks (DNN) to estimate crowd counts via density regression. Despite their significant improvements, the regression-based methods are incapable of providing the detection of individuals in crowds. The detection-based methods, on the other hand, have not been largely explored in recent trends of crowd counting due to the needs for expensive bounding box annotations. In this work, we instead propose a new deep detection network with only point supervision required. It can simultaneously detect the size and location of human heads and count them in crowds. We first mine useful person size information from pointlevel annotations and initialize the pseudo ground truth bounding boxes. An online updating scheme is introduced to refine the pseudo ground truth during training; while a locally-constrained regression loss is designed to provide additional constraints on the size of the predicted boxes in a local neighborhood. In the end, we propose a curriculum learning strategy to train the network from images of relatively accurate and easy pseudo ground truth first. Extensive experiments are conducted in both detection and counting tasks on several standard benchmarks, e.g. Shang-haiTech, UCF CC 50, WiderFace, and TRANCOS datasets, and the results show the superiority of our method over the state-of-the-art. arXiv:1904.01333v2 [cs.CV]

No abstract

This paper proposes an encoder-decoder network to disentangle shape features during 3D face reconstruction from single 2D images, such that the tasks of reconstructing accurate 3D face shapes and learning discriminative shape features for face recognition can be accomplished simultaneously. Unlike existing 3D face reconstruction methods, our proposed method directly regresses dense 3D face shapes from single 2D images, and tackles identity and residual (i.e., non-identity) components in 3D face shapes explicitly and separately based on a composite 3D face shape model with latent representations. We devise a training process for the proposed network with a joint loss measuring both face identification error and 3D face shape reconstruction error. To construct training data we develop a method for fitting 3D morphable model (3DMM) to multiple 2D images of a subject. Comprehensive experiments have been done on MICC, BU3DFE, LFW and YTF databases. The results show that our method expands the capacity of 3DMM for capturing discriminative shape features and facial detail, and thus outperforms existing methods both in 3D face reconstruction accuracy and in face recognition accuracy.

Abstract. Sweat pores on fingerprints have proven to be useful features for personal identification. Several methods have been proposed for pore matching. The state-of-the-art method first matches minutiae on the fingerprints and then matches the pores based on the minutia matching results. A problem of such minutia-based pore matching method is that the pore matching is dependent on the minutia matching. Such dependency limits the pore matching performance and impairs the effectiveness of the fusion of minutia and pore match scores. In this paper, we propose a novel direct approach for matching fingerprint pores. It first determines the correspondences between pores based on their local features. It then uses the RANSAC (RANdom SAmple Consensus) algorithm to refine the pore correspondences obtained in the first step. A similarity score is finally calculated based on the pore matching results. The proposed pore matching method successfully avoids the dependency of pore matching on minutia matching results. Experiments have shown that the fingerprint recognition accuracy can be greatly improved by using the method proposed in this paper.

No abstract

No abstract

No abstract