The 2016 IEEE GRSS Data Fusion Contest, organized by the IADF TC, was opened on January 3, 2016. The submission deadline was April 29, 2016. Participants submitted open topic manuscripts using the VHR and video-from-space data released for the competition.
25 teams worldwide participated to the Contest. Evaluation and ranking were conducted by the Award Committee. The winners are reported below along with the abstracts of the submitted papers.
Title: Spatiotemporal scene interpretation of space videos via deep neural network and tracklet analysis
Authors: Lichao Mou and Xiaoxiang Zhu
Abstract: Spaceborne remote sensing videos are becoming indispensable resources, opening up opportunities for new remote sensing applications. To exploit this new type of data, we need sophisticated algorithms for semantic scene interpretation. The main difficulties are: 1) Due to the relatively poor spatial resolution of the video acquired from space, moving objects, like cars, are very difficult to detect, not to mention track; 2) Camera movement handicaps scene interpretation. To address these challenges, in this paper we propose a novel framework that fuses multispectral images and space videos for spatiotemporal analysis. Taking a multispectral image and a spaceborne video as input, an innovative deep neural network is proposed to fuse them in order to achieve a fine-resolution spatial scene labeling map. Moreover, a sophisticated approach is proposed to analyze activities and estimate traffic density from 150; 000+ tracklets produced by a Kanade-Lucas-Tomasi keypoint tracker. The proposed framework is validated using data provided for the 2016 IEEE GRSS data fusion contest, including a video acquired from the International Space Station and a DEIMOS-2 multispectral image. Both visual and quantitative analysis of the experimental results demonstrates the effectiveness of our approach.
Title:Simultaneous registration, segmentation and change detection from multisensor, multitemporal satellite image pairs
Authors: Maria Vakalopoulou, Christos Platias, Maria Papadomanolaki, Nikos Paragios and Konstantinos Karantzalos
Abstract: In this paper, a novel framework has been designed, developed and validated for addressing simultaneously the tasks of image registration, segmentation and change detection from multisensor, multiresolution, multitemporal satellite image pairs. Our approach models the inter-dependencies of variables through a higher order graph. The proposed formulation is modular with respect to the nature of images (various similarity metrics can be considered), the nature of deformations (arbitrary interpolation strategies), and the nature of segmentation likelihoods (various classification approaches can be employed). Inference of the proposed formulation is achieved through its mapping to an over-parametrized pairwise graph which is then optimized using linear programming. Experimental results and the performed quantitative evaluation indicate the high potentials of the developed method.
Title:Sensor-agnostic photogrammetric image registration with applications to population modeling
Authors: Dave Kelbe, Devin White, Andrew Hardin, Jessica Moehl and Melanie Phillips.
Abstract: While wide area motion imagery provides short-timescale temporal information, e.g., individual vehicle tracking, it lacks broader contextual information on the ambient distribution of populations within that area. We present a fusion approach to augment Iris video with broader-scale population data. Spectral, geometric, and geospatial limitations of the Iris video preclude the use of Iris video directly; this is overcome by photogrammetric registration of robust Deimos-2 imagery and ancillary processed products using a high performance sensor-agnostic, multi-temporal registration workflow. We assess the accuracy and precision of the proposed workflow (~15 m; Euclidean) and demonstrate the potential to leverage the fusion of these data towards rapid, global-scale population distribution modeling. This has important implications to effective response to emergencies, especially in urban environments, where population density is driven largely by building heights, and a complementary, multi-scale understanding of the distribution and dynamics of people within that geographic area is required.
Title:Building extraction from multi-source remote sensing images via deep deconvolution neural networks
Authors:Zuming Huang, Guangliang Cheng, Hongzhen Wang, Haichang Li, Limin Shi and Chunhong Pan.
Abstract: Building extraction from remote sensing images is of great importance in urban planing. Yet it is a longstanding problem for many complicate factors such as various scales and complex backgrounds. This paper proposes a novel supervised building extraction method via deep deconvolution neural networks (DeconvNet). Our method consists of three steps. First, we preprocess the multi-source remote sensing images provided by the IEEE GRSS Data Fusion Contest. A high-quality Vancouver building dataset is created on pansharpened images whose ground-truth are obtained from the OpenStreetMap project. Then, we pretrain a deep deconvolution network on a public large-scale Massachusetts building dataset, which is further fine-tuned by two band combinations (RGB and NRG) of our dataset respectively. Moreover, the output saliency maps of the fine-tuned models are fused to produce the final building extraction result. Extensive experiments on our Vancouver building dataset demonstrate the effectiveness and efficiency of the proposed method. To the best of our knowledge, it is the first work to use deconvolution networks for building extraction from remote sensing images.