Cross-view geo-localization provides an offline visual positioning strategy for unmanned aerial vehicles (UAVs) in Global Navigation Satellite System (GNSS)-denied environments. However, it still faces the following challenges, leading to suboptimal localization performance: 1) Existing methods primarily focus on extracting global features or local features by partitioning feature maps, neglecting the exploration of spatial information, which is essential for extracting consistent feature representations and aligning images of identical targets across different views. 2) Cross-view geo-localization encounters the challenge of data imbalance between UAV and satellite images. To address these challenges, the Spatial Hybrid Attention Network with Adaptive Cross-Entropy Loss Function (SHAA) is proposed. To tackle the first issue, the Spatial Hybrid Attention (SHA) method employs a Spatial Shift-MLP (SSM) to focus on the spatial geometric correspondences in feature maps across different views, extracting both global features and fine-grained features. Additionally, the SHA method utilizes a Hybrid Attention (HA) mechanism to enhance feature extraction diversity and robustness by capturing interactions between spatial and channel dimensions, thereby extracting consistent cross-view features and aligning images. For the second challenge, the Adaptive Cross-Entropy (ACE) loss function incorporates adaptive weights to emphasize hard samples, alleviating data imbalance issues and improving training effectiveness. Extensive experiments on widely recognized benchmarks, including University-1652, SUES-200, and DenseUAV, demonstrate that SHAA achieves state-of-the-art performance, outperforming existing methods by over 3.92%.

The all-to-all collective communications primitive is widely used in machine learning (ML) and high performance computing (HPC) workloads, and optimizing its performance is of interest to both ML and HPC communities. All-to-all is a particularly challenging workload that can severely strain the underlying interconnect bandwidth at scale. This paper takes a holistic approach to optimize the performance of all-to-all collective communications on supercomputer-scale direct-connect interconnects. We address several algorithmic and practical challenges in developing efficient and bandwidth-optimal all-to-all schedules for any topology and lowering the schedules to various runtimes and interconnect technologies. We also propose a novel topology that delivers near-optimal all-to-all performance.

For indoor environments, prior point-based visual SLAM cannot be processed in real time under low texture and illumination. To address this issue, this work proposes PLP-SLAM (Point-Line-Plane-SLAM) with RGB-D camera. Firstly, point and line features are detected in RGB images. For line features, establish length suppression and near line merge strategy to improve the line extraction quality. Secondly, plane features are extracted based on agglomerative hierarchical clustering method in point cloud obtained by RGB-D camera. Point clouds are divided into several nodes, unlike prior methods spend a lot of time to estimate the normal vector for each individual point, this work assumes that points within each node sharing the same plane normal vector, which can significantly improve the computational efficiency. Thirdly, sparse maps including points, lines and planes are established, meanwhile the scenes are reconstructed by creating the dense maps to show plan features directly. Finally, the performance of proposed method is compared against the state-of-the-art SLAM on public datasets to evaluate the pose estimation. All modules are run in real-time on a CPU, experiments clarify that PLP-SLAM can significantly enhance the robustness of 6DoF pose of the camera and simultaneously creating more detailed maps of the environment.

Cloud operators utilize collective communication optimizers to enhance the efficiency of the single-tenant, centrally managed training clusters they manage. However, current optimizers struggle to scale for such use cases and often compromise solution quality for scalability. Our solution, TE-CCL, adopts a traffic-engineering-based approach to collective communication. Compared to a state-of-the-art optimizer, TACCL, TE-CCL produced schedules with 2× better performance on topologies TACCL supports (and its solver took a similar amount of time as TACCL's heuristic-based approach). TECCL additionally scales to larger topologies than TACCL. On our GPU testbed, TE-CCL outperformed TACCL by 2.14× and RCCL by 3.18× in terms of algorithm bandwidth.

We address the problem of serving Deep Neural Networks (DNNs) efficiently from a cluster of GPUs. In order to realize the promise of very low-cost processing made by accelerators such as GPUs, it is essential to run them at sustained high utilization. Doing so requires cluster-scale resource management that performs detailed scheduling of GPUs, reasoning about groups of DNN invocations that need to be co-scheduled, and moving from the conventional whole-DNN execution model to executing fragments of DNNs. Nexus is a fully implemented system that includes these innovations. In large-scale case studies on 16 GPUs, when required to stay within latency constraints at least 99% of the time, Nexus can process requests at rates 1.8-12.7X higher than state of the art systems can. A long-running multi-application deployment stays within 84% of optimal utilization and, on a 100-GPU cluster, violates latency SLOs on 0.27% of requests.