Segmenting Sky Pixels in Images

Outdoor scene parsing models are often trained on ideal datasets and produce quality results. However, this leads to a discrepancy when applied to the real world. The quality of scene parsing, particularly sky classification, decreases in night time images, images involving varying weather conditions, and scene changes due to seasonal weather. This project focuses on approaching these challenges by using a state-of-the-art model in conjunction with a non-ideal dataset: SkyFinder and a subset from SUN database with Sky object. We focus specifically on sky segmentation, the task of determining sky and not-sky pixels, and improving upon an existing state-of-the-art model: RefineNet. As a result of our efforts, we have seen an improvement of 10-15% in the average MCR compared to the prior methods on SkyFinder dataset. We have also improved from an off-the shelf-model in terms of average mIOU by nearly 35%. Further, we analyze our trained models on images w.r.t two aspects: times of day and weather, and find that, in spite of facing same challenges as prior methods, our trained models significantly outperform them.

[1]  F. Moosmann,et al.  Classification of weather situations on single color images , 2008, 2008 IEEE Intelligent Vehicles Symposium.

[2]  Svetlana Lazebnik,et al.  Superparsing , 2010, International Journal of Computer Vision.

[3]  Svetlana Lazebnik,et al.  Finding Things: Image Parsing with Regions and Per-Exemplar Detectors , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Stefan Winkler,et al.  Color-Based Segmentation of Sky/Cloud Images From Ground-Based Cameras , 2016, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[5]  Antonio Torralba,et al.  Nonparametric scene parsing: Label transfer via dense scene alignment , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  R. Ruijsink,et al.  Sky Segmentation Approach to obstacle avoidance , 2011, 2011 Aerospace Conference.

[7]  Krista A. Ehinger,et al.  SUN database: Large-scale scene recognition from abbey to zoo , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[8]  Xiaogang Wang,et al.  Pyramid Scene Parsing Network , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Emma E. Regentova,et al.  Sky Segmentation by Fusing Clustering with Neural Networks , 2013, ISVC.

[10]  Alexei A. Efros,et al.  Geometric context from a single image , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[11]  Ming-Hsuan Yang,et al.  Sky is not the limit , 2016, ACM Trans. Graph..

[12]  Xiaofeng Tao,et al.  Transient attributes for high-level understanding and editing of outdoor scenes , 2014, ACM Trans. Graph..

[13]  Ian D. Reid,et al.  RefineNet: Multi-path Refinement Networks for High-Resolution Semantic Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Wei-Ta Chu,et al.  Camera as weather sensor: Estimating weather information from single images , 2017, J. Vis. Commun. Image Represent..

[15]  Scott Workman,et al.  Sky segmentation in the wild: An empirical study , 2016, 2016 IEEE Winter Conference on Applications of Computer Vision (WACV).

[16]  Svetlana Lazebnik,et al.  Scene Parsing with Object Instances and Occlusion Ordering , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  Xiaoming Zheng,et al.  Weather Recognition Based on Images Captured by Vision System in Vehicle , 2009, ISNN.

[18]  Jian Sun,et al.  SkyFinder: attribute-based sky image search , 2009, ACM Trans. Graph..

[19]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).