1,721,164 research outputs found
Web-Scale Image Retrieval and Its Novel Applications
No full textContent-based image retrieval has been studied more than two decades, but surprisingly has not been widely deployed for web-scale image databases. This is mainly caused by two factors: low scalability and lack of commercial applications. In this poster we review our recent research activities for developing web-scalable image retrieval and its novel applications to address various problems. First we discuss scalability issues to be addressed for handling web-scale image databases. We also describe our hashing technique to drastically lower down the memory and computational requirements for web-scale image retrieval. Second we explain its applications to the fields of image copyright and editing, and discuss benefits that it brings over prior techniques developed without considering or utilizing web-scale image databases. Finally, we discuss remained problems towards our goal
Enhancing Navigation Efficiency of Quadruped Robots via Leveraging Personal Transportation Platforms
No full textRegional Attention Based Deep Feature for Image Retrieval
No full textMany approaches using Convolutional Neural Network (CNN) for efficient image retrieval have concentrated on feature aggregation rather than feature embedding over recent years, since convolutional features have been found to be reasonably discriminative. Nonetheless, we found that a well-known region-based feature aggregation method, R-MAC, for image retrieval is suffered from the background clutter and varying importance of regions. In this work, we tackle these problems with a simple and effective, context-aware regional attention network that weights an attentive score of a region considering global attentiveness. We conduct various experiments on well-known retrieval datasets, and confirm that our method does not only improve the R-MAC baseline significantly, but also present new state-of-the-art results in the category of ``pre-trained single-pass’’. Furthermore, we show that our method shows higher accuracy improvement combined over prior methods, when combined with the query expansion method. These results are attributed by our novel regional-attention network integrated with R-MA
Cache-efficient layouts of bounding volume hierarchies
No full textWe present a novel algorithm to compute cache-efficient layouts of bounding volume hierarchies (BVHs) of polygonal models. Our approach does not make any assumptions about the cache parameters or block sizes of the memory hierarchy. We introduce a new probabilistic model to predict the runtime access patterns of a BVH. Our layout computation algorithm utilizes parent-child and spatial localities between the accessed nodes to reduce both the number of cache misses and the size of the working set. Our algorithm also works well for spatial partitioning hierarchies including kd-trees. We use our algorithm to compute layouts of BVHs and spatial partitioning hierarchies of large models composed of millions of triangles. We compare our cache-efficient layouts with other layouts in the context of collision detection and ray tracing. In our benchmarks, our layouts consistently show better performance over other layouts and improve the performance of these applications by 26%-300% without any modification of the underlying algorithms or runtime applications.We would like to thank Ivy Bigelow, Russ Gayle, Dawoon
Jung, Ted Kim, Ming Lin, Peter Lindstrom, Brandon Lloyd,
Valerio Pascucci, Stephane Redon, and anonymous reviewers
for useful discussions and feedback. The Bunny, Dragon, and
Lucy models are courtesy of Stanford University. The power
plant model is courtesy of an anonymous donor. This work
was supported in part by ARO Contracts DAAD19-02-1-0390
and W911NF-04-1-0088, NSF awards 0400134 and 0118743,
ONR Contract N00014-01-1-0496, DARPA/RDECOM Contract
N61339-04-C-0043, Intel, and, LOCAL LLNL LDRD
project (05-ERD-018). Some of the work was performed under
the auspices of the U.S. Department of Energy by the University
of California, Lawrence Livermore National Laboratory
under Contract No. W-7405-Eng-48
Harmonious Sampling for Mobile Manipulation Planning
No full textMobile manipulation planning commonly adopts a decoupled approach that performs planning separately on the base and the manipulator. While this approach is fast, it can generate sub-optimal paths. Another direction is a coupled approach jointly adjusting the base and manipulator in a high-dimensional configuration space. This coupled approach addresses sub-optimality and incompleteness of the decoupled approach, but has not been widely used due to its excessive computational overhead. Given this trade-off space, we present a simple, yet effective mobile manipulation sampling method, harmonious sampling, to perform the coupled approach mainly in difficult regions, where we need to simultaneously maneuver the base and the manipulator. Our method identifies such difficult regions through a low-dimensional base space by utilizing a reachability map given the target end-effector pose and narrow passage detected by generalized Voronoi diagram. For the rest of simple regions, we perform sampling mainly on the base configurations with a predefined joint configuration, accelerating the planning process. We compare our method with the decoupled and coupled approaches in six different problems with varying difficulty. Our method shows meaningful improvements experimentally in terms of time to find an initial solution (up to 5.6 times faster) and final solution cost (up to 17% lower) over the decoupled approach, especially in difficult scenes with narrow space. We also demonstrate these benefits with a real, mobile Hubo robot
PROT: Productive Regions-Oriented Task space path planning for hyper-redundant manipulators
No full textGroup Estimation for Social Robot Navigation in Crowded Environments
No full textSocially acceptable navigation in a crowded environment is a challenging problem in robotics due to diverse and unknown human intent. Previous studies have dealt with the social navigation problem in dense crowds via multi-robot collision avoidance. However, it is intractable to follow social compliant trajectory since human-robot interaction differs from the multi-robot collision avoidance problem. To approach our goal, this work exploits a human behavior model and focuses on social group actions such as walking together. We observed that human recognizes the other human groups and avoids them during navigation while maintaining social distances. Based on this observation, this paper proposes a social robot navigation method under group space estimation of crowds on a deep reinforcement learning framework. The proposed method estimates the social groups of crowds based on the behavioral similarities in sensory information. Our reinforcement learning framework learns a socially compliant and effective navigation policy through the proposed human group-aware reward. Our experiment in a crowd simulation demonstrates that the proposed approach generates a human-friendly trajectory with improved navigation performance
Recent Advances in Real-Time Collision andProximity Computations for Games and Simulations
No full text- …
