SM-NAS: Structural-to-Modular Neural Architecture Search for Object Detection

By Lewei Yao, Hang Xu, Wei Zhang, Xiaodan Liang, Zhenguo Li. 2019

86 citations recorded by the source corpus.

The state-of-the-art object detection method is complicated with various modules such as backbone, RPN, feature fusion neck and RCNN head, where each module may have different designs and structures. How to leverage the computational cost and accuracy trade-off for the structural combination as well as the modular selection of multiple modules? Neural architecture search (NAS) has shown great potential in finding an optimal solution. Existing NAS works for object detection only focus on searching better design of a single module such as backbone or feature fusion neck, while neglecting the balance of the whole system. In this paper, we present a two-stage coarse-to-fine searching strategy named Structural-to-Modular NAS (SM-NAS) for searching a GPU-friendly design of both an efficient combination of modules and better modular-level architecture for object detection. Specifically, Structural-level searching stage first aims to find an efficient combination of different modules; Modular-level searching stage then evolves each specific module and pushes the Pareto front forward to a faster task-specific network. We consider a multi-objective search where the search space covers many popular designs of detection methods. We directly search a detection backbone without pre-trained models or any proxy task by exploring a fast training from scratch strategy. The resulting architectures dominate state-of-the-art object detection systems in both inference time and accuracy and demonstrate the effectiveness on multiple detection datasets, e.g. halving the inference time with additional 1% mAP improvement compared to FPN and reaching 46% mAP with the similar inference time of MaskRCNN.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

YOLOv4: Optimal Speed and Accuracy of Object Detection (2020, background): “SM-NAS: Structural-to-Modular Neural Architecture Search for Object Detection [88] SM-NAS: E2 800×600 25.”
SODA10M: Towards Large-Scale Object Detection Benchmark for Autonomous Driving (2021, methodology): “Pre-annotation: In order to ensure efficiency, a multi-task detection model, which is based on Faster RCNN [45] and searched backbone [29, 62], is trained on millions of Human-Vehicle images with bounding-box annotation and generate coarse labels for each image first.”
CurveLane-NAS: Unifying Lane-Sensitive Architecture Search and Adaptive Point Blending (2020, background): “Most works search a basic CNN architectures for a classification model [18, 3, 16, 26, 29] while a few of them focus on more complicated high-level vision tasks such as semantic segmentation and object detection [4, 5, 30, 31].”