In the pursuit of excellence in competitive sports, improving the performance of track and field athletes is a continuous pursuit. The rise of computer big data technology has brought new breakthroughs to sports train...
详细信息
LiDAR and cameras are two widely used sensors in 3D object detection. LiDAR point clouds show geometry knowledge of objects, while RGB images provide semantic information, such as color and texture. How to effectively...
详细信息
ISBN:
(纸本)9798350352634;9798350352627
LiDAR and cameras are two widely used sensors in 3D object detection. LiDAR point clouds show geometry knowledge of objects, while RGB images provide semantic information, such as color and texture. How to effectively fuse their features is the key to improving detection performance. This paper proposes a Deformable Feature Fusion Network, which performs LiDAR-camera fusion in a flexible way. We present multi-modal features in the bird's-eye view(BEV), and build a Deformable-Attention Fusion(DAF) module to conduct feature fusion. Besides fusion methods, feature alignment is also important in multi-modal detection. Data augmentation of point clouds may change the projection relationship between RGB images and LiDAR point clouds and causes feature misalignment. We introduce a Feature Alignment Transform(FAT) module and alleviate the problem without introducing any trainable parameters. We conduct experiments on the KITTI dataset to evaluate the effectiveness of proposed modules and the experiment results show that our method outperforms most existing methods.
When solving the problem of path planning method for mobile robot obstacle avoidance, the newly proposed Dingo optimization algorithm (DOA) is used for path planning, and the efficiency of path planning is improved by...
详细信息
The current manual approach to assessing external damage to submarine cables lacks standardized risk identification and grading standards, and falls short in providing comprehensive and continuous risk monitoring. To ...
详细信息
5G networks has advanced connectivity, high data, low latency and reliability. Along with these features new challenges come with respect to security, particularly spoofing attacks. One of the necessity in 5G and futu...
详细信息
To solve the problems of lengthy computation times and slow path searches for the 5-DOF cooperative robot, an algorithm based on the long short-term memory network (LSTM) is proposed in this paper. Firstly, the forwar...
详细信息
This paper proposes a mobile application that translates vocal input into real-time, editable arithmetic equations, thus serving as a flexible input method for mathematical expressions. The application will allow user...
详细信息
This study explored the factors influencing organizations39; artificial intelligence and machine learning technology adoption in the context of MSMEs of developing countries with particular reference to the Indian c...
详细信息
The inaccuracy of modeling information and external disturbance bring great challenges to the control of robot manipulators. In the paper, an adaptive control strategy of robot manipulators with model uncertainty is p...
详细信息
AHRS-LeGO-LOAM: Based on the LeGO-LOAM algorithm (a lightweight ground-optimized Lidar ranging and mapping method), we have added a new measurement module IMU (AHRS Attitude and Heading Reference System) with magnetom...
详细信息
暂无评论