VIO-40K

Tianxia Liu, Bofeng Li, Guang’e Chen, Ling Yang, Jing Qiao, Wu Chen

The technology of autonomous vehicle (AV) is critical in nowadays Intelligent Transportation Systems. To achieve the fully automated operation for AVs, one important prerequisite is the accurate and reliable seamless localization covering complex outdoor-indoor scenarios. Although many solutions have been proposed to support AV localization, it is still challenging in achieving reliable drift-free positioning in seamless urban environments. With the current on-board sensors such as GNSS, IMU, LiDAR and cameras, it is difficult to achieve accurate drift-free indoor positioning due to the lack of GNSS indoors. Meanwhile, challenges remain in reliable navigation under obscured conditions. In this paper, we propose a tightly coupled integration algorithm of GNSS RTK, Ultra-Wide Band (UWB) and Visual Inertial Odometry (VIO) to enhance the accuracy and reliability for AVs seamless localization in challenging environments. The UWB technique is innovatively incorporated into the AVs navigation system to extend absolute positioning indoors. The stereo cameras are utilized to improve positioning continuity and enhance GNSS/UWB usability in outdoor-indoor obscured environments. The proposed algorithm is evaluated over real-world datasets in complex seamless environments. The results show that the proposed algorithm achieves 0.411m and 0.077m horizontal positioning accuracy in obscured outdoor and indoor environments, yielding 71.2% and 18.1% improvements compared with the traditional LC integration schemes, respectively.

Shuhao Han, Haotian Fan, Jiachen Fu, Liang Li, Tao Li, Junhui Cui, Yunqiu Wang, Yang Tai, Jingwei Sun, Chunle Guo, Chongyi Li

Recently, Text-to-Image (T2I) generation models have achieved significant advancements. Correspondingly, many automated metrics have emerged to evaluate the image-text alignment capabilities of generative models. However, the performance comparison among these automated metrics is limited by existing small datasets. Additionally, these datasets lack the capacity to assess the performance of automated metrics at a fine-grained level. In this study, we contribute an EvalMuse-40K benchmark, gathering 40K image-text pairs with fine-grained human annotations for image-text alignment-related tasks. In the construction process, we employ various strategies such as balanced prompt sampling and data re-annotation to ensure the diversity and reliability of our benchmark. This allows us to comprehensively evaluate the effectiveness of image-text alignment metrics for T2I models. Meanwhile, we introduce two new methods to evaluate the image-text alignment capabilities of T2I models: FGA-BLIP2 which involves end-to-end fine-tuning of a vision-language model to produce fine-grained image-text alignment scores and PN-VQA which adopts a novel positive-negative VQA manner in VQA models for zero-shot fine-grained evaluation. Both methods achieve impressive performance in image-text alignment evaluations. We also use our methods to rank current AIGC models, in which the results can serve as a reference source for future study and promote the development of T2I generation. The data and code will be made publicly available.

M. Siraz, S. Das, M. S. Mondol, M. Alam, Jubair Al Mahmud, Md. Bazlar Rashid, Mayeen Uddin Khandaker, S. Yeasmin

Youqi Pan, Wugen Zhou, Yingdian Cao, Hongbin Zha

Visual-inertial odometry (VIO) has demonstrated re-markable success due to its low-cost and complementary sensors. However, existing VIO methods lack the general-ization ability to adjust to different environments and sen-sor attributes. In this paper, we propose Adaptive VIO, a new monocular visual-inertial odometry that combines online continual learning with traditional nonlinear opti-mization. Adaptive VIO comprises two networks to pre-dict visual correspondence and IMU bias. Unlike end-to-end approaches that use networks to fuse the features from two modalities (camera and IMU) and predict poses directly, we combine neural networks with visual-inertial bundle adjustment in our VIO system. The optimized esti-mates will be fed back to the visual and IMU bias networks, refining the networks in a self-supervised manner. Such a learning-optimization-combined framework and feedback mechanism enable the system to perform online contin-ual learning. Experiments demonstrate that our Adaptive VIO manifests adaptive capability on EuRoC and TUM-VI datasets. The overall performance exceeds the currently known learning-based VIO methods and is comparable to the state-of-the-art optimization-based methods.

J. Desimoni, F. Sives, L. Errico, G. Mastrantonio, M. Taylor