Image-to-point-cloud (I2P) registration is a fundamental yet challenging problem in computer vision. Despite significant advances in deep learning, I2P registration struggles with correspondence accuracy when training samples are limited. To address this challenge, we propose a Beltrami flow based I2P registration method termed Flow-I2P. From the perspective of information geometry, I2P registration can be reframed as a manifold alignment problem. Our in-depth analysis shows that Beltrami flow enhances I2P registration by improving manifold alignment quality. Building on this analysis, we introduce a Beltrami flow based cross-modality feature interaction layer, B-flow, to progressively refine manifold alignment. To reduce memory and computation demands, B-flow is then optimized into C-flow through the incorporation of feature covariance-based attention. We further enhance I2P registration performance by developing Flow-I2P, which incorporates normal features, stacked C-flow layers, and a two-stage training strategy. To evaluate the registration performance of Flow-I2P, we conduct extensive experiments on five indoor and outdoor datasets, including RGB-D V2, 7-Scenes, ScanNet, KITTI, and a self-collected dataset. Our results indicate that Flow-I2P achieves higher inlier ratio (IR) and registration recall (RR) compared to state-of-the-art methods. We conclude that Flow-I2P significantly enhances I2P registration with superior capabilities.

The Japanese Quasi-Zenith Satellite System (QZSS) satellite system has placed in orbit four satellites by October 2017. The Indian Regional Navigation Satellite System (IRNSS) system has launched the new satellite IRNNSS-11 in April 2018, completing seven operational satellites. Together with the GPS block IIF satellites and the Galileo satellites, four different global navigation satellite systems (GNSSs) are providing precise L5 signals on the frequency of 1176.45 MHz. In this contribution, we challenge the strength of the multi-GNSS model by analysing its single-frequency (L5), single-epoch (instantaneous) precise positioning capabilities under high-elevation masking (up to 40 degrees). With more satellites available, multi-GNSS real time kinematic (RTK) positioning is possible using L5-only signals with a high customary elevation mask. This helps to enable positioning in areas with constrained measurement geometry, and could significantly reduce the multipath effects in difficult measurement environments like urban canyons and mountainous areas. In this study, benefiting from the location of the Asia⁻Australia area, instantaneous multi-GNSS L5 RTK analysis is performed with respect to the ambiguity resolution and positioning performance. Formal results are shown and discussed for baselines located in different grids covering Australia, part of the Pacific Ocean, Indian Ocean and Asia, and empirical analysis is given for two baselines in Perth, Australia. Compared to the stand-alone cases, for baselines in Perth, it is shown that combining L5 signals from GPS/Galileo/QZSS/IRNSS significantly improves both the ambiguity success rates (ASR) and the positioning performance under high elevation mask. While the average single-system ASR is under 50% even with a low elevation mask of 10 degrees, combining all the four systems increases the ASR to above 95% under an elevation cut-off angles of 40 degrees. With an elevation mask of 40 degrees, using satellites from one system does not allow for meaningful positioning solutions of more than 8 h within the test day, while mm-to-cm level ambiguity-fixed standard deviations could be obtained based on the positioning results of almost the entire day when combining all the four systems. In addition to that, simulation was also performed for receivers with larger signal standard deviations, i.e., for low-cost receivers or receivers located in environments with larger multipath.

In this contribution we analyze the single-frequency L5 positioning capabilities of the two regional satellite navigation systems IRNSS and QZSS, stand alone as well as combined. The positioning analysis is done for two different baselines, having a mix of receivers, providing ambiguity-float and ambiguity-fixed positioning for models with and without zenith tropospheric delay (ZTD) estimation. The analyses include a precision analysis of the observed signals, as well as an analysis of the ambiguity resolution performance. This is done for both the multipath-uncorrected case as well as the multipath-mitigated case. It is shown that although single-system positioning performance is rather poor, the ZTD-fixed, single-epoch ambiguity success rates (ASRs) are close to 100% when the two regional systems are combined, thus providing mm-to-cm level precision for instantaneous ambiguity-fixed positioning. When the ZTD is estimated as well, only a few additional epochs are needed to get the ASRs close to 100%.

In this contribution, we study the phase-only ambiguity resolution and positioning performance of GPS for short baselines. It is well known that instantaneous (single-epoch) ambiguity resolution is possible when both phase and code (pseudorange) data are used. This requires, however, a benign multipath environment due to the severe effects multipath has on the code measurements. With phase-only processing, one would be free from such severe effects, be it that phase-only processing requires a change in receiver-satellite geometry, as a consequence of which it cannot be done instantaneously. It is thus of interest to know how much change in the relative receiver-satellite geometry is needed to achieve successful phase-only ambiguity resolution with correspondingly high precision baseline solutions. In this contribution, we study the two-epoch phase-only performance of single-, dual-, and triple-frequency GPS for varying time spans from 60 s down to 1 s. We demonstrate, empirically as well as formally, that fast phase-only very-precise positioning is indeed possible, and we explain the circumstances that make this possible. The formal analyses are also performed for a large area including Australia, a part of Asia, the Indian Ocean, and the Pacific Ocean. We remark that in this contribution "phase-only" refers to phase-only measurements in the observation model, while the code data are thus only used to compute the approximate values needed for linearizing the observation equations.