Small-sized unmanned surface vehicles (USV) are coastal water devices with a broad range of applications such as environmental control and surveillance. A crucial capability for autonomous operation is obstacle detection for timely reaction and collision avoidance, which has been recently explored in the context of camera-based visual scene interpretation. Owing to curated datasets, substantial advances in scene interpretation have been made in a related field of unmanned ground vehicles. However, the current maritime datasets do not adequately capture the complexity of real-world USV scenes and the evaluation protocols are not standardised, which makes cross-paper comparison of different methods difficult and hinders the progress. To address these issues, we introduce a new obstacle detection benchmark MODS, which considers two major perception tasks: maritime object detection and the more general maritime obstacle segmentation. We present a new diverse maritime evaluation dataset containing approximately 81k stereo images synchronized with an on-board IMU, with over 60k objects annotated. We propose a new obstacle segmentation performance evaluation protocol that reflects the detection accuracy in a way meaningful for practical USV navigation. Nineteen recent state-of-the-art object detection and obstacle segmentation methods are evaluated using the proposed protocol, creating a benchmark to facilitate development of the field. The proposed dataset, as well as evaluation routines, are made publicly available at vicos.si/resources.

In this paper, we present a method for detecting and tracking waterborne obstacles from an unmanned surface vehicle (USV) for the purpose of short-term obstacle avoidance. A stereo camera system provides a point cloud of the scene in front of the vehicle. The water surface is estimated by fitting a plane to the point cloud and outlying points are further processed to find potential obstacles. We propose a new plane fitting algorithm for water surface detection that applies a fast approximate semantic segmentation to filter the point cloud and utilizes an external IMU reading to constrain the plane orientation. A novel histogram-like depth appearance model is proposed to keep track of the identity of the detected obstacles through time and to filter out false detections that negatively impact the vehicle's automatic guidance system. The improved plane fitting algorithm and the temporal verification using depth fingerprints result in notable improvement on the challenging MODD2 data set by significantly reducing the amount of false positive detections. The proposed method is able to run in real time on board of a small-sized USV, which was used to acquire the MODD2 data set as well.

Object grasping is a fundamental challenge in robotics and computer vision, critical for advancing robotic manipulation capabilities. Deformable objects, like fabrics and cloths, pose additional challenges due to their non-rigid nature. In this work, we introduce CeDiRNet-3DoF, a deep-learning model for grasp point detection, with a particular focus on cloth objects. CeDiRNet-3DoF employs center direction regression alongside a localization network, attaining first place in the perception task of ICRA 2023's Cloth Manipulation Challenge. Recognizing the lack of standardized benchmarks in the literature that hinder effective method comparison, we present the ViCoS Towel Dataset. This extensive benchmark dataset comprises 8,000 real and 12,000 synthetic images, serving as a robust resource for training and evaluating contemporary data-driven deep-learning approaches. Extensive evaluation revealed CeDiRNet-3DoF's robustness in real-world performance, outperforming state-of-the-art methods, including the latest transformer-based models. Our work bridges a crucial gap, offering a robust solution and benchmark for cloth grasping in computer vision and robotics. Code and dataset are available at: https://github.com/vicoslab/CeDiRNet-3DoF .

Camera systems in autonomous vehicles are subject to various sources of anticipated and unanticipated mechanical stress (vibration, rough handling, collisions) in real-world conditions. Even moderate changes in camera geometry due to mechanical stress decalibrate multi-camera systems and corrupt downstream applications like depth perception. We propose an on-the-fly stereo recalibration method applicable in real-world autonomous vehicles. The method is comprised of two parts. First, in optimization step, external camera parameters are optimized with the goal to maximise the amount of recovered depth pixels. In the second step, external sensor is used to adjust the scaling of the optimized camera model. The method is lightweight and fast enough to run in parallel with stereo estimation, thus allowing an on-the-fly recalibration. Our extensive experimental analysis shows that our method achieves stereo reconstruction better or on par with manual calibration. If our method is used on a sequence of images, the quality of calibration can be improved even further.

Multimodal sensor systems require precise calibration if they are to be used in the field. Due to the difficulty of obtaining the corresponding features from different modalities, the calibration of such systems is an open problem. We present a systematic approach for calibrating a set of cameras with different modalities (RGB, thermal, polarization, and dual-spectrum near infrared) with regard to a LiDAR sensor using a planar calibration target. Firstly, a method for calibrating a single camera with regard to the LiDAR sensor is proposed. The method is usable with any modality, as long as the calibration pattern is detected. A methodology for establishing a parallax-aware pixel mapping between different camera modalities is then presented. Such a mapping can then be used to transfer annotations, features, and results between highly differing camera modalities to facilitate feature extraction and deep detection and segmentation methods.

We address the problem of optical decalibration in mobile stereo camera setups, especially in context of autonomous vehicles. In real world conditions, an optical system is subject to various sources of anticipated and unanticipated mechanical stress (vibration, rough handling, collisions). Mechanical stress changes the geometry between the cameras that make up the stereo pair, and as a consequence, the pre-calculated epipolar geometry is no longer valid. Our method is based on optimization of camera geometry parameters and plugs directly into the output of the stereo matching algorithm. Therefore, it is able to recover calibration parameters on image pairs obtained from a decalibrated stereo system with minimal use of additional computing resources. The number of successfully recovered depth pixels is used as an objective function, which we aim to maximize. Our simulation confirms that the method can run constantly in parallel to stereo estimation and thus help keep the system calibrated in real time. Results confirm that the method is able to recalibrate all the parameters except for the baseline distance, which scales the absolute depth readings. However, that scaling factor could be uniquely determined using any kind of absolute range finding methods (e.g. a single beam time-of-flight sensor).