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.
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.