FasterSeg: Searching for Faster Real-time Semantic Segmentation

Wuyang Chen, Xinyu Gong, Xianming Liu, Qian Zhang, Yuan Li, Zhangyang Wang

Keywords: neural architecture search, real-time, segmentation

TL;DR: We present a real-time segmentation model automatically discovered by a multi-scale NAS framework, achieving 30% faster than state-of-the-art models.

Abstract: We present FasterSeg, an automatically designed semantic segmentation network with not only state-of-the-art performance but also faster speed than current methods. Utilizing neural architecture search (NAS), FasterSeg is discovered from a novel and broader search space integrating multi-resolution branches, that has been recently found to be vital in manually designed segmentation models. To better calibrate the balance between the goals of high accuracy and low latency, we propose a decoupled and fine-grained latency regularization, that effectively overcomes our observed phenomenons that the searched networks are prone to "collapsing" to low-latency yet poor-accuracy models. Moreover, we seamlessly extend FasterSeg to a new collaborative search (co-searching) framework, simultaneously searching for a teacher and a student network in the same single run. The teacher-student distillation further boosts the student model’s accuracy. Experiments on popular segmentation benchmarks demonstrate the competency of FasterSeg. For example, FasterSeg can run over 30% faster than the closest manually designed competitor on Cityscapes, while maintaining comparable accuracy.

Code: https://github.com/TAMU-VITA/FasterSeg

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:1912.10917/code)