Associative scene semantic segmentation장면 분할을 위한 연상 도메인 적응 기법

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 312
  • Download : 0
This paper considers an associative unsupervised domain adaptation learning algorithm for performing semantic segmentation on real urban drive-cam data using photo-realistic synthetic training data. To circumvent the difficulty of collecting and laboriously annotating a large amounts of real urban scene data, large amounts of computer-annotated synthetic training data is provided as a substitute; however, without any consideration to domain mismatch, a significant decreases in prediction performance is observed. Inspired by the recent success of an associative domain adaptation algorithm for simple classification, this algorithm is adapted to semantic segmentation to reduce domain mismatch between training and testing. Considering associative learning for multiple instances within a single high-resolution image and ambiguous and undecided labels in a semantic segmentation training dataset, this adaptation is not straightforward. In this paper, an algorithm is proposed to address such difficulties in adapting associative learning to semantic segmentation by partitioning an image into patches and associating labeled patches with unlabeled patches. The results from the model using SYNTHIA and GTA5 dataset as a source dataset shows state-of-the-art performance on the CityScapes dataset.
Advisors
Yoo, ChangDongresearcher유창동researcher
Description
한국과학기술원 :전기및전자공학부,
Publisher
한국과학기술원
Issue Date
2018
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2018.2,[iii, 21 p. :]

Keywords

semantic segmentation▼aunsupervised learning▼adomain adaptation▼asynthetic to real▼ascene segmentation; 영상분할▼a비지도식 도메인 적응▼a연상▼a딥려닝▼a기계학습

URI
http://hdl.handle.net/10203/266854
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=734068&flag=dissertation
Appears in Collection
EE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0