Gunho Sohn

Sunghwan Yoo, Yeongjeong Jeong, Maryam Jameela et al.

This paper proposes EyeNet, a novel semantic segmentation network for point clouds that addresses the critical yet often overlooked parameter of coverage area size. Inspired by human peripheral vision, EyeNet overcomes the limitations of conventional networks by introducing a simple but efficient multi-contour input and a parallel processing network with connection blocks between parallel streams. The proposed approach effectively addresses the challenges of dense point clouds, as demonstrated by our ablation studies and state-of-the-art performance on Large-Scale Outdoor datasets.

Jungwon Kang, Mohammadjavad Ghorbanalivakili, Gunho Sohn et al.

One essential feature of an autonomous train is minimizing collision risks with third-party objects. To estimate the risk, the control system must identify topological information of all the rail routes ahead on which the train can possibly move, especially within merging or diverging rails. This way, the train can figure out the status of potential obstacles with respect to its route and hence, make a timely decision. Numerous studies have successfully extracted all rail tracks as a whole within forward-looking images without considering element instances. Still, some image-based methods have employed hard-coded prior knowledge of railway geometry on 3D data to associate left-right rails and generate rail route instances. However, we propose a rail path extraction pipeline in which left-right rail pixels of each rail route instance are extracted and associated through a fully convolutional encoder-decoder architecture called TPE-Net. Two different regression branches for TPE-Net are proposed to regress the locations of center points of each rail route, along with their corresponding left-right pixels. Extracted rail pixels are then spatially clustered to generate topological information of all the possible train routes (ego-paths), discarding non-ego-path ones. Experimental results on a challenging, publicly released benchmark show true-positive-pixel level average precision and recall of 0.9207 and 0.8721, respectively, at about 12 frames per second. Even though our evaluation results are not higher than the SOTA, the proposed regression pipeline performs remarkably in extracting the correspondences by looking once at the image. It generates strong rail route hypotheses without reliance on camera parameters, 3D data, and geometrical constraints.

Mostafa Ahmadi, Amin Alizadeh Naeini, Mohammad Moein Sheikholeslami et al.

This paper proposes a novel visual simultaneous localization and mapping (SLAM) system called Hybrid Depth-augmented Panoramic Visual SLAM (HDPV-SLAM), that employs a panoramic camera and a tilted multi-beam LiDAR scanner to generate accurate and metrically-scaled trajectories. RGB-D SLAM was the design basis for HDPV-SLAM, which added depth information to visual features. It aims to solve the two major issues hindering the performance of similar SLAM systems. The first obstacle is the sparseness of LiDAR depth, which makes it difficult to correlate it with the extracted visual features of the RGB image. A deep learning-based depth estimation module for iteratively densifying sparse LiDAR depth was suggested to address this issue. The second issue pertains to the difficulties in depth association caused by a lack of horizontal overlap between the panoramic camera and the tilted LiDAR sensor. To surmount this difficulty, we present a hybrid depth association module that optimally combines depth information estimated by two independent procedures, feature-based triangulation and depth estimation. During a phase of feature tracking, this hybrid depth association module aims to maximize the use of more accurate depth information between the triangulated depth with visual features tracked and the deep learning-based corrected depth. We evaluated the efficacy of HDPV-SLAM using the 18.95 km-long York University and Teledyne Optech (YUTO) MMS dataset. The experimental results demonstrate that the two proposed modules contribute substantially to the performance of HDPV-SLAM, which surpasses that of the state-of-the-art (SOTA) SLAM systems.

Mohammad Koushafar, Gunho Sohn, Mark Gordon

Estimating Plume Cloud (PC) height is essential for various applications, such as global climate models. Smokestack Plume Rise (PR) is the constant height at which the PC is carried downwind as its momentum dissipates and the PC and the ambient temperatures equalize. Although different parameterizations are used in most air-quality models to predict PR, they have yet to be verified thoroughly. This paper proposes a low-cost measurement technology to monitor smokestack PCs and make long-term, real-time measurements of PR. For this purpose, a two-stage method is developed based on Deep Convolutional Neural Networks (DCNNs). In the first stage, an improved Mask R-CNN, called Deep Plume Rise Network (DPRNet), is applied to recognize the PC. Here, image processing analyses and least squares, respectively, are used to detect PC boundaries and fit an asymptotic model into the boundaries centerline. The y-component coordinate of this model's critical point is considered PR. In the second stage, a geometric transformation phase converts image measurements into real-life ones. A wide range of images with different atmospheric conditions, including day, night, and cloudy/foggy, have been selected for the DPRNet training algorithm. Obtained results show that the proposed method outperforms widely-used networks in smoke border detection and recognition.

Maryam Jameela, Gunho Sohn

Due to the aged nature of much of the utility network infrastructure, developing a robust and trustworthy computer vision system capable of inspecting it with minimal human intervention has attracted considerable research attention. The airborne laser terrain mapping (ALTM) system quickly becomes the central data collection system among the numerous available sensors. Its ability to penetrate foliage with high-powered energy provides wide coverage and achieves survey-grade ranging accuracy. However, the post-data acquisition process for classifying the ALTM's dense and irregular point clouds is a critical bottleneck that must be addressed to improve efficiency and accuracy. We introduce a novel deep convolutional neural network (DCNN) technique for achieving voxel-based semantic segmentation of the ALTM's point clouds. The suggested deep learning method, Semantic Utility Network (SUNet) is a multi-dimensional and multi-resolution network. SUNet combines two networks: one classifies point clouds at multi-resolution with object categories in three dimensions and another predicts two-dimensional regional labels distinguishing corridor regions from non-corridors. A significant innovation of the SUNet is that it imposes spatial layout consistency on the outcomes of voxel-based and regional segmentation results. The proposed multi-dimensional DCNN combines hierarchical context for spatial layout embedding with a coarse-to-fine strategy. We conducted a comprehensive ablation study to test SUNet's performance using 67 km x 67 km of utility corridor data at a density of 5pp/m2. Our experiments demonstrated that SUNet's spatial layout consistency and a multi-resolution feature aggregation could significantly improve performance, outperforming the SOTA baseline network and achieving a good F1 score for pylon 89%, ground 99%, vegetation 99% and powerline 98% classes.

Maryam Jameela, Gunho Sohn

LiDAR (Light Detection and Ranging) technology has remained popular in capturing natural and built environments for numerous applications. The recent technological advancements in electro-optical engineering have aided in obtaining laser returns at a higher pulse repetition frequency (PRF), which considerably increased the density of the 3D point cloud. Conventional techniques with lower PRF had a single pulse-in-air (SPIA) zone, large enough to avoid a mismatch among pulse pairs at the receiver. New multiple pulses-in-air (MPIA) technology guarantees various windows of operational ranges for a single flight line and no blind zones. The disadvantage of the technology is the projection of atmospheric returns closer to the same pulse-in-air zone of adjacent terrain points likely to intersect with objects of interest. These noise properties compromise the perceived quality of the scene and encourage the development of new noise-filtering neural networks, as existing filters are significantly ineffective. We propose a novel dual-attention noise-filtering neural network called Noise Seeking Attention Network (NSANet) that uses physical priors and local spatial attention to filter noise. Our research is motivated by two psychology theories of feature integration and attention engagement to prove the role of attention in computer vision at the encoding and decoding phase. The presented results of NSANet show the inclination towards attention engagement theory and a performance boost compared to the state-of-the-art noise-filtering deep convolutional neural networks.

Mohammad Moein Sheikholeslami, Muhammad Kamran, Andreas Wichmann et al.

The growing demand for high-resolution maps across various applications has underscored the necessity of accurately segmenting building vectors from overhead imagery. However, current deep neural networks often produce raster data outputs, leading to the need for extensive post-processing that compromises the fidelity, regularity, and simplicity of building representations. In response, this paper introduces a novel deep convolutional neural network named OriCornerNet, which directly extracts delineated building polygons from input images. Specifically, our approach involves a deep model that predicts building footprint masks, corners, and orientation vectors that indicate directions toward adjacent corners. These predictions are then used to reconstruct an initial polygon, followed by iterative refinement using a graph convolutional network that leverages semantic and geometric features. Our method inherently generates simplified polygons by initializing the refinement process with predicted corners. Also, including geometric information from oriented corners contributes to producing more regular and accurate results. Performance evaluations conducted on SpaceNet Vegas and CrowdAI-small datasets demonstrate the competitive efficacy of our approach compared to the state-of-the-art in building segmentation from overhead imagery.

Muhammad Kamran, Mohammad Moein Sheikholeslami, Andreas Wichmann et al.

In recent years, the number of remote satellites orbiting the Earth has grown significantly, streaming vast amounts of high-resolution visual data to support diverse applications across civil, public, and military domains. Among these applications, the generation and updating of spatial maps of the built environment have become critical due to the extensive coverage and detailed imagery provided by satellites. However, reconstructing spatial maps from satellite imagery is a complex computer vision task, requiring the creation of high-level object representations, such as primitives, to accurately capture the built environment. While the past decade has witnessed remarkable advancements in object detection and representation using visual data, primitives-based object representation remains a persistent challenge in computer vision. Consequently, high-quality spatial maps often rely on labor-intensive and manual processes. This paper introduces a novel deep learning methodology leveraging Graph Convolutional Networks (GCNs) to address these challenges in building footprint reconstruction. The proposed approach enhances performance by incorporating geometric regularity into building boundaries, integrating multi-scale and multi-resolution features, and embedding Attraction Field Maps into the network. These innovations provide a scalable and precise solution for automated building footprint extraction from a single satellite image, paving the way for impactful applications in urban planning, disaster management, and large-scale spatial analysis. Our model, Decoupled-PolyGCN, outperforms existing methods by 6% in AP and 10% in AR, demonstrating its ability to deliver accurate and regularized building footprints across diverse and challenging scenarios.

Kang Zhao, Muhammad Kamran, Gunho Sohn

In recent years, an ever-increasing number of remote satellites are orbiting the Earth which streams vast amount of visual data to support a wide range of civil, public and military applications. One of the key information obtained from satellite imagery is to produce and update spatial maps of built environment due to its wide coverage with high resolution data. However, reconstructing spatial maps from satellite imagery is not a trivial vision task as it requires reconstructing a scene or object with high-level representation such as primitives. For the last decade, significant advancement in object detection and representation using visual data has been achieved, but the primitive-based object representation still remains as a challenging vision task. Thus, a high-quality spatial map is mainly produced through complex labour-intensive processes. In this paper, we propose a novel deep neural network, which enables to jointly detect building instance and regularize noisy building boundary shapes from a single satellite imagery. The proposed deep learning method consists of a two-stage object detection network to produce region of interest (RoI) features and a building boundary extraction network using graph models to learn geometric information of the polygon shapes. Extensive experiments show that our model can accomplish multi-tasks of object localization, recognition, semantic labelling and geometric shape extraction simultaneously. In terms of building extraction accuracy, computation efficiency and boundary regularization performance, our model outperforms the state-of-the-art baseline models.