In this paper, we introduce a new large-scale dataset of ships, called SeaShips, which is designed for training and evaluating ship object detection algorithms. The dataset currently consists of 31 455 images and covers six common ship types (ore carrier, bulk cargo carrier, general cargo ship, container ship, fishing boat, and passenger ship). All of the images are from about 10 080 real-world video segments, which are acquired by the monitoring cameras in a deployed coastline video surveillance system. They are carefully selected to mostly cover all possible imaging variations, for example, different scales, hull parts, illumination, viewpoints, backgrounds, and occlusions. All images are annotated with ship-type labels and high-precision bounding boxes. Based on the SeaShips dataset, we present the performance of three detectors as a baseline to do the following: 1) elementarily summarize the difficulties of the dataset for ship detection; 2) show detection results for researchers using the dataset; and 3) make a comparison to identify the strengths and weaknesses of the baseline algorithms. In practice, the SeaShips dataset would hopefully advance research and applications on ship detection.

Most SLAM (Simultaneous Localization and Mapping) algorithms are based on the assumption that the scene is static. However, in practice, most real scenes usually contain moving objects. In this letter, we introduce DymSLAM, a dynamic stereo visual SLAM system being capable of reconstructing a 4D (3D + time) dynamic scene with rigid moving objects. Unlike previous attempts that have considered moving objects as outliers and ignored them, DymSLAM obtains the 6DoF motion trajectory and 3D models about the dynamic objects. We segment motion models of different moving objects by a multi-motion segmentation approach and obtain the accurate masks of moving objects. Besides ego-motion, our system can obtain the 4D (3D + time) model and 6DoF trajectory of the moving object in the global reference frame while simultaneously reconstructing the dense map of the static background. Meanwhile, DymSLAM does not rely on semantic cues or prior knowledge and is suitable for unknown rigid objects. We conducted experiments in a real-world indoor environment where both the camera and the objects were moving in a wide range. The results proved that our proposed method is a state-of-the-art SLAM system for use in this dynamic environment.

Different from object detection in natural image, optical remote sensing object detection is a challenging task, due to the diverse meteorological conditions, complex background, varied orientations, scale variations, etc. In this paper, to address this issue, we propose a novel object detection network (the global-local saliency constraint network, GLS-Net) that can make full use of the global semantic information and achieve more accurate oriented bounding boxes. More precisely, to improve the quality of the region proposals and bounding boxes, we first propose a saliency pyramid which combines a saliency algorithm with a feature pyramid network, to reduce the impact of complex background. Based on the saliency pyramid, we then propose a global attention module branch to enhance the semantic connection between the target and the global scenario. A fast feature fusion strategy is also used to combine the local object information based on the saliency pyramid with the global semantic information optimized by the attention mechanism. Finally, we use an angle-sensitive intersection over union (IoU) method to obtain a more accurate five-parameter representation of the oriented bounding boxes. Experiments with a publicly available object detection dataset for aerial images demonstrate that the proposed GLS-Net achieves a state-of-the-art detection performance.

The integration of the camera and LiDAR has played an important role in the field of autonomous driving, for example in visual–LiDAR SLAM and 3D environment fusion perception, which rely on precise geometrical extrinsic calibration. In this paper, we proposed a fully automatic end-to-end method based on the 3D–2D corresponding mask (CoMask) to directly estimate the extrinsic parameters with high precision. Simple subtraction was applied to extract the candidate point cluster from the complex background, and then 3D LiDAR points located on checkerboard were selected and refined by spatial growth clustering. Once the distance transform of 2D checkerboard mask was generated, the extrinsic calibration of the two sensors could be converted to 3D–2D mask correspondence alignment. A simple but efficient strategy combining the genetic algorithm with the Levenberg–Marquardt method was used to solve the optimization problem globally without any initial estimates. Both simulated and realistic experiments showed that the proposed method could obtain accurate results without manual intervention, special environment setups, or prior initial parameters. Compared with the state of the art, our method has obvious advantages in accuracy, robustness, and noise resistance. Our code is open-source on GitHub.

Most scenes in practical applications are dynamic scenes containing moving objects, so accurately segmenting moving objects is crucial for many computer vision applications. In order to efficiently segment all the moving objects in the scene, regardless of whether the object has a predefined semantic label, we propose a two-level nested octave U-structure network with a multi-scale attention mechanism, called U²-ONe$\mathrm{t}$. ${\mathrm{U}}^2$-ONe$\mathrm{t}$ takes two RGB frames, the optical flow between these frames, and the instance segmentation of the frames as inputs. Each stage of ${\mathrm{U}}^2$-ONe$\mathrm{t}$ is filled with the newly designed octave residual U-block (ORSU block) to enhance the ability to obtain more contextual information at different scales while reducing the spatial redundancy of the feature maps. In order to efficiently train the multi-scale deep network, we introduce a hierarchical training supervision strategy that calculates the loss at each level while adding knowledge-matching loss to keep the optimization consistent. The experimental results show that the proposed ${\mathrm{U}}^2$-ONe$\mathrm{t}$ method can achieve a state-of-the-art performance in several general moving object segmentation datasets.