English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Semantic segmentation of road scenes is one of the key technologies forrealizing autonomous driving scene perception, and the effectiveness of deepConvolutional Neural Networks(CNNs) for this task has been demonstrated.State-of-art CNNs for semantic segmentation suffer from excessive computationsas well as large-scale training data requirement. Inspired by the ideas ofFine-tuning-based Transfer Learning (FTT) and feature-based knowledgedistillation, we propose a new knowledge distillation method for cross-domainknowledge transference and efficient data-insufficient network training, namedSpirit Distillation(SD), which allow the student network to mimic the teachernetwork to extract general features, so that a compact and accurate studentnetwork can be trained for real-time semantic segmentation of road scenes.Then, in order to further alleviate the trouble of insufficient data andimprove the robustness of the student, an Enhanced Spirit Distillation (ESD)method is proposed, which commits to exploit a more comprehensive generalfeatures extraction capability by considering images from both the target andthe proximity domains as input. To our knowledge, this paper is a pioneeringwork on the application of knowledge distillation to few-shot learning.Persuasive experiments conducted on Cityscapes semantic segmentation with theprior knowledge transferred from COCO2017 and KITTI demonstrate that ourmethods can train a better student network (mIOU and high-precision accuracyboost by 1.4% and 8.2% respectively, with 78.2% segmentation variance) withonly 41.8% FLOPs (see Fig. 1).

Spirit Distillation: Precise Real-time Semantic Segmentation of Road Scenes with Insufficient Data