Multi-object 3D Grounding involves locating 3D boxes based on a given query phrase from a point cloud. It is a challenging and significant task that has numerous applications in visual understanding, human-computer interaction, and robotics. To tackle this challenge, we introduce D-LISA, a two-stage approach that incorporates three innovations. First, a dynamic vision module that enables a variable and learnable number of box proposals. Second, a dynamic camera positioning that extracts features for each proposal. Third, a language-informed spatial attention module that better reasons over the proposals to output the final prediction. Empirically, experiments show that our method outperforms the state-of-the-art methods on multi-object 3D grounding by 12.8% (absolute) and is competitive in single-object 3D grounding.

Most real-world 3D measurements from depth sensors are incomplete, and to address this issue the point cloud completion task aims to predict the complete shapes of objects from partial observations. Previous works often adapt an encoder-decoder architecture, where the encoder is trained to extract embeddings that are used as inputs to generate predictions from the decoder. However, the learned embeddings have sparse distribution in the feature space, which leads to worse generalization results during testing. To address these problems, this paper proposes a hyperspherical module, which transforms and normalizes embeddings from the encoder to be on a unit hypersphere. With the proposed module, the magnitude and direction of the output hyperspherical embedding are decoupled and only the directional information is optimized. We theoretically analyze the hyperspherical embedding and show that it enables more stable training with a wider range of learning rates and more compact embedding distributions. Experiment results show consistent improvement of point cloud completion in both single-task and multi-task learning, which demonstrates the effectiveness of the proposed method.

This paper considers the single source shortest path (SSSP) problem, which is the key for many applications such as navigation, mapping, routing, and social networking. Existing SSSP algorithms are designed mostly for shared-memory systems. Nevertheless, with the prevalence of diverse smart devices like drones, there is a growing interest in deploying SSSP algorithms over distributed computing systems so that they can run efficiently onboard of smart devices via Mobile Ad Hoc Computing or at the network edges via Mobile Edge Computing. In this paper, we introduce a communication-efficient ∆-stepping algorithm for distributed computing systems. The proposed algorithm is featured by 1) a message coordination architecture for reducing message exchanges between workers, 2) a pruning technique for reducing redundant computations, and 3) an aggregation technique for further reducing message exchanges when communication delay is significant. Theoretical analyses and experimental studies on real-world graph datasets demonstrate the promising performance of proposed algorithm.