2020
DOI: 10.48550/arxiv.2004.02147
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

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

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
71
0
1

Year Published

2021
2021
2022
2022

Publication Types

Select...
6
3

Relationship

0
9

Authors

Journals

citations
Cited by 46 publications
(72 citation statements)
references
References 51 publications
0
71
0
1
Order By: Relevance
“…To further improve the segmentation accuracy, we propose a booster training strategy [24], called the Split Soul (SS) module. This module consists of a 3 × 3 global average pooling layer, a 1 × 1 convolution layer, and a 3 × 3 convolution layer.…”
Section: Split Soul (Ss) Modulementioning
confidence: 99%
“…To further improve the segmentation accuracy, we propose a booster training strategy [24], called the Split Soul (SS) module. This module consists of a 3 × 3 global average pooling layer, a 1 × 1 convolution layer, and a 3 × 3 convolution layer.…”
Section: Split Soul (Ss) Modulementioning
confidence: 99%
“…Following previous works [32,45,51], we perform semantic segmentation on synthesized images to quantify how well the predicted segments match ground-truth. Specifically, BiSeNet [46] is applied to the synthesized images to infer semantic segmentation results, and pixelwise accuracy (pix acc) and mean intersection-over-union (mIoU) are utilized as its evaluation metrics. Furthermore, we compare SuperStyleNet with these state-of-the-art methods by adopting peak signal-to-noise ratio (PSNR), normalized root mean square error (NRMSE), Fréchet Inception Distance (FID) [15], and learned perceptual image patch similarity (LPIPS) [49].…”
Section: Evaluation Metricsmentioning
confidence: 99%
“…Semantic segmentation is a classic and fundamental topic in computer vision, which aims to assign pixellevel labels in images. The prosperity of deep learning greatly promotes the performance of semantic segmentation by making various breakthroughs [18,27,22,4], coming with fast-growing demands in many applications, e.g., autonomous driving, video surveillance, robot sensing, and so on. These applications motivate researchers to explore effective and efficient segmentation networks, particularly for mobile field.…”
Section: Introductionmentioning
confidence: 99%
“…
BiSeNet [28,27] has been proved to be a popular twostream network for real-time segmentation. However, its principle of adding an extra path to encode spatial information is time-consuming, and the backbones borrowed from pretrained tasks, e.g., image classification, may be inefficient for image segmentation due to the deficiency of taskspecific design.
…”
mentioning
confidence: 99%