Semantic segmentation methods have made impressive progress with deep learning. However, while achieving higher and higher accuracy, state-of-the-art neural networks overlook the complexity of architectures, which typically feature dozens of millions of trainable parameters. Consequently, these networks requires high computational ressources and are mostly not suited to perform on edge devices with tight resource constraints, such as phones, drones, or satellites. In this work, we propose two highlycompact neural network architectures for semantic segmentation of images, which are up to 100 000 times less complex than state-of-the-art architectures while approaching their accuracy. To decrease the complexity of existing networks, our main ideas consist in exploiting lightweight encoders and decoders with depth-wise separable convolutions and decreasing memory usage with the removal of skip connections between encoder and decoder. Our architectures are designed to be implemented on a basic FPGA such as the one featured on the Intel Altera Cyclone V family of SoCs. We demonstrate the potential of our solutions in the case of binary segmentation of remote sensing images, in particular for extracting clouds and trees from RGB satellite images.
Over the past decades, 3D face has emerged as a solution to face recognition due to its reputed invariance to lighting conditions and pose. While proposed approaches have proven their efficiency over renowned databases as FRGC, less effort was spent on studying the robustness of algorithms to quality degradations. In this paper, we present a study of the robustness of four state of the art algorithms and a multi-matcher framework to face model degradations such as Gaussian noise, decimation, and holes. The four state of the art algorithms were chosen for their different and complementary properties and exemplify the major classes of 3D face recognition solutions. As they displayed different behavior under data degradations, we further designed a fusion framework to best take into account their complementary properties. The proposed multi-matcher scheme is based on an offline and an online weight learning process. Experiments were conducted on a subset of the FRGC database, on which we generated degradations. Results demonstrate the competitive robustness of the proposed approach.
While object detection methods traditionally make use of pixel-level masks or bounding boxes, alternative representations such as polygons or active contours have recently emerged. Among them, methods based on the regression of Fourier or Chebyshev coefficients have shown high potential on freeform objects. By defining object shapes as polar functions, they are however limited to star-shaped domains. We address this issue with SCR: a method that captures resolution-free object contours as complex periodic functions. The method offers a good compromise between accuracy and compactness thanks to the design of efficient geometric shape priors. We benchmark SCR on the popular COCO 2017 instance segmentation dataset, and show its competitiveness against existing algorithms in the field. In addition, we design a compact version of our network, which we benchmark on embedded hardware with a wide range of power targets, achieving up to real-time performance.
Digital data acquisition and control systems present challenges that many of today's computer systems are not equipped to address. Many of the performance requirements in a digital data acquisition and control system may be met by an intelligent device capable of precise, high speed operations. A solution for those performance requirements is presented here with a dual processor board providing flexibility in both hardware and software. The processor is capable of executing over 12 million instructions per second, serves as its own development system, and runs an interpretive, high level programming language. Included is a discussion of the application of this processor in a radar data acquisition system.
In this paper, a demonstrator called BIOFACE incorporating several facial biometric techniques is described. It includes the well established Eigenfaces and the recently published Tomofaces techniques, which perform face recognition based on facial appearance and dynamics, respectively. Both techniques are based on the space dimensionality reduction and the enrollment requires the projection of several positive face samples to the reduced space. Alternatively, BIOFACE also performs face recognition based on the matching of Scale Invariant Feature Transform (SIFT) features.Moreover, BIOFACE extracts a facial soft biometric profile, which consists of a bag of facial soft biometric traits such as skin, hair, and eye color, the presence of glasses, beard and moustache. The fast and efficient detection of the facial soft biometrics is performed as a pre-processing step, and employed for pruning the search for the facial recognition module.Finally, the demonstrator also detects facial events such as blinking, yawning and looking-away. The car driver scenario is a good example to exhibit the importance of such traits to detect fatigue.The BIOFACE demonstrator is an attempt to show the potential and the performance of such facial processing techniques in a real-life scenario. The demonstrator is built using the C/C++ programming language, which is suitable for implementing image and video processing techniques due to its fast execution. On top of that, the Open Source Computer Vision Library (OpenCV), which is optimized for Intel processors, is used to implement the image processing algorithms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.