BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation

by   Changqian Yu, et al.

The low-level details and high-level semantics are both essential to the semantic segmentation task. However, to speed up the model inference, current approaches almost always sacrifice the low-level details, which leads to a considerable accuracy decrease. We propose to treat these spatial details and categorical semantics separately to achieve high accuracy and high efficiency for realtime semantic segmentation. To this end, we propose an efficient and effective architecture with a good trade-off between speed and accuracy, termed Bilateral Segmentation Network (BiSeNet V2). This architecture involves: (i) a Detail Branch, with wide channels and shallow layers to capture low-level details and generate high-resolution feature representation; (ii) a Semantic Branch, with narrow channels and deep layers to obtain high-level semantic context. The Semantic Branch is lightweight due to reducing the channel capacity and a fast-downsampling strategy. Furthermore, we design a Guided Aggregation Layer to enhance mutual connections and fuse both types of feature representation. Besides, a booster training strategy is designed to improve the segmentation performance without any extra inference cost. Extensive quantitative and qualitative evaluations demonstrate that the proposed architecture performs favourably against a few state-of-the-art real-time semantic segmentation approaches. Specifically, for a 2,048x1,024 input, we achieve 72.6 one NVIDIA GeForce GTX 1080 Ti card, which is significantly faster than existing methods, yet we achieve better segmentation accuracy.


page 10

page 11

page 14


CSDN: Combing Shallow and Deep Networks for Accurate Real-time Segmentation of High-definition Intravascular Ultrasound Images

Intravascular ultrasound (IVUS) is the preferred modality for capturing ...

Bi-direction Context Propagation Network for Real-time Semantic Segmentation

Spatial details and context correlations are two types of critical infor...

PIDNet: A Real-time Semantic Segmentation Network Inspired from PID Controller

Two-branch network architecture has shown its efficiency and effectivene...

SLLEN: Semantic-aware Low-light Image Enhancement Network

How to effectively explore semantic feature is vital for low-light image...

Highly Efficient Follicular Segmentation in Thyroid Cytopathological Whole Slide Image

In this paper, we propose a novel method for highly efficient follicular...

Rethinking BiSeNet For Real-time Semantic Segmentation

BiSeNet has been proved to be a popular two-stream network for real-time...

DANCE: DAta-Network Co-optimization for Efficient Segmentation Model Training and Inference

Semantic segmentation for scene understanding is nowadays widely demande...

Please sign up or login with your details

Forgot password? Click here to reset