Customizable Architecture Search for Semantic Segmentation

Abstract

In this paper, we propose a Customizable Architecture Search (CAS) approach to automatically generate a network architecture for semantic image segmentation. The generated network consists of a sequence of stacked computation cells. A computation cell is represented as a directed acyclic graph, in which each node is a hidden representation (i.e., feature map) and each edge is associated with an operation (e.g., convolution and pooling), which transforms data to a new layer. During the training, the CAS algorithm explores the search space for an optimized computation cell to build a network. The cells of the same type share one architecture but with different weights. In real applications, however, an optimization may need to be conducted under some constraints such as GPU time and model size. To this end, a cost corresponding to the constraint will be assigned to each operation. When an operation is selected during the search, its associated cost will be added to the objective. As a result, our CAS is able to search an optimized architecture with customized constraints. The approach has been thoroughly evaluated on Cityscapes and CamVid datasets, and demonstrates superior performance over several stateof-the-art techniques. More remarkably, our CAS achieves 72.3% mIoU on the Cityscapes dataset with speed of 108 FPS on an Nvidia TitanXp GPU.

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

“…architecture depends on the difficulty and size of the dataset at hand. While these findings may encourage an automated neural architecture search, such an approach is hindered by the limited computational resources [19], [20], [21], [22], [23]. Alternatively, we propose an ensemble architecture, which combines U-Nets of varying depths into one unified structure.…”

Searching Central Difference Convolutional Networks for Face Anti-Spoofing

“…Most of NAS approaches search networks on a small proxy task and transfer the found architecture to another large target task. For the perspective of computer vision applications, NAS has been developed for face recognition [67], action recognition [46], person ReID [50], object detection [21] and segmentation [65] tasks. To the best of our knowledge, no NAS based method has ever been proposed for face anti-spoofing task.…”

3D-Bioprinting Strategies Based on In Situ Bone-Healing Mechanism for Vascularized Bone Tissue Engineering

“…Recently, 3D-bioprinting technologies have emerged to provide practical solutions in various medical fields to fabricate biomimetic multicellular tissues with a highly complex microenvironment, such as combining 3D-printing, tissue engineering, developmental biology, and regenerative medicine [ 123 ]. Similar to 3D-printing, as an additive manufacturing methodology, 3D-bioprinting can precisely control the complex 3D architectures, spatial distributions, and positioning of multiple compositions in a layer-by-layer manner to deposit biomaterials (called bioinks as they allow printing of living cells), in which cells and signaling cues may be embedded as customizing patient-specific therapies [ 124 ].…”

“…Extrusion-based bioprinting is the most common, affordable, and easy-to-use bioprinting approach among the three classified bioprinting technologies [ 129 ]. Bioinks with physiological cell densities can be extruded through microscale nozzles using precisely controlled pneumatic pressure or mechanical compressions within a temperature-controlled system [ 124 , 126 ]. Crosslinking of bioinks can be achieved during dispensing and/or depositing processes by chemical, thermal, and/or photocurable means according to the type of polymer [ 126 ].…”