Skip to Main content Skip to Navigation

Semantic Segmentation of Highly Structured and Weakly Structured Images

Abstract : The aim of this thesis is to develop techniques for segmenting strongly-structuredscenes (e.g. building images) and weakly-structured scenes (e.g. natural images). Buildingimages can naturally be expressed in terms of grammars and inference is performed usinggrammars to obtain the optimal segmentation. However, it is difficult and time consum-ing to write such grammars. To alleviate this problem, a novel method to automaticallylearn grammars from a given training set of image and ground-truth segmentation pairs isdeveloped. Experiments suggested that such learned grammars help in better and fasterinference. Next, the effect of using grammars for strongly structured scenes is explored.To this end, a very simple technique based on Auto-Context is used to segment buildingimages. Surprisingly, even with out using any domain specific knowledge, we observedsignificant improvements in terms of performance on several benchmark datasets. Lastly,a novel technique based on convolutional neural networks is developed to segment imageswithout any high-level structure. Image-adaptive filtering is performed within a CNN ar-chitecture to facilitate long-range connections. Experiments on different large scale bench-marks show significant improvements in terms of performance
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Monday, March 26, 2018 - 6:08:11 PM
Last modification on : Saturday, January 15, 2022 - 3:50:10 AM
Long-term archiving on: : Thursday, September 13, 2018 - 9:29:29 AM


Version validated by the jury (STAR)


  • HAL Id : tel-01743925, version 1


Raghu Deep Gadde. Semantic Segmentation of Highly Structured and Weakly Structured Images. Signal and Image Processing. Université Paris-Est, 2017. English. ⟨NNT : 2017PESC1083⟩. ⟨tel-01743925⟩



Record views


Files downloads