arXiv subjects

Kazuhiro Shintani

Publications and source records attributed to Kazuhiro Shintani.

5 recordsLinked to original sources

arXiv2026

Where Do Humans Look When Demonstrating to Robots? Human Gaze Behavior in Pick-and-Place Tasks Across Demonstration Devices

Imitation learning for generalizable performance often requires a large volume of demonstration data, making the process significantly costly. One promising strategy to address this challenge is to leverage the cognitive skills of human demonstrators with strong generalization capability, particularly by revealing the underlying task demands reflected in their gaze behavior. However, imitation learning typically involves humans collecting data using demonstration devices that emulate a robot's embodiment and visual condition. This raises the question of how such devices influence gaze behavior. We propose an experimental framework that systematically analyzes human demonstrators' gaze behavior across a spectrum of robot-emulating demonstration devices. Our experimental results show that certain device properties shift gaze from task-goal cues (e.g., objects) toward control-monitoring cues (e.g., the end-effector). Furthermore, these shifts directly affect the performance of typical gaze-based imitation learning models, sometimes degrading it below non-gaze baselines.

cs.RO↗

arXiv2025

Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry

Monocular visual odometry is a key technology in various autonomous systems. Traditional feature-based methods suffer from failures due to poor lighting, insufficient texture, and large motions. In contrast, recent learning-based dense SLAM methods exploit iterative dense bundle adjustment to address such failure cases, and achieve robust and accurate localization in a wide variety of real environments, without depending on domain-specific supervision. However, despite its potential, the methods still struggle with scenarios involving large motion and object dynamics. In this study, we diagnose key weaknesses in a popular learning-based dense SLAM model (DROID-SLAM) by analyzing major failure cases on outdoor benchmarks and exposing various shortcomings of its optimization process. We then propose the use of self-supervised priors leveraging a frozen large-scale pre-trained monocular depth estimator to initialize the dense bundle adjustment process, leading to robust visual odometry without the need to fine-tune the SLAM backbone. Despite its simplicity, the proposed method demonstrates significant improvements on KITTI odometry, as well as the challenging DDAD benchmark.

cs.CV↗

arXiv2024

CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision Language Models for Object-based Global Localization

This letter proposes a method of global localization on a map with semantic object landmarks. One of the most promising approaches for localization on object maps is to use semantic graph matching using landmark descriptors calculated from the distribution of surrounding objects. These descriptors are vulnerable to misclassification and partial observations. Moreover, many existing methods rely on inlier extraction using RANSAC, which is stochastic and sensitive to a high outlier rate. To address the former issue, we augment the correspondence matching using Vision Language Models (VLMs). Landmark discriminability is improved by VLM embeddings, which are independent of surrounding objects. In addition, inliers are estimated deterministically using a graph-theoretic approach. We also incorporate pose calculation using the weighted least squares considering correspondence similarity and observation completeness to improve the robustness. We confirmed improvements in matching and pose estimation accuracy through experiments on ScanNet and TUM datasets.

cs.CV↗

arXiv2024

Robust Imitation Learning for Mobile Manipulator Focusing on Task-Related Viewpoints and Regions

We study how to generalize the visuomotor policy of a mobile manipulator from the perspective of visual observations. The mobile manipulator is prone to occlusion owing to its own body when only a single viewpoint is employed and a significant domain shift when deployed in diverse situations. However, to the best of the authors' knowledge, no study has been able to solve occlusion and domain shift simultaneously and propose a robust policy. In this paper, we propose a robust imitation learning method for mobile manipulators that focuses on task-related viewpoints and their spatial regions when observing multiple viewpoints. The multiple viewpoint policy includes attention mechanism, which is learned with an augmented dataset, and brings optimal viewpoints and robust visual embedding against occlusion and domain shift. Comparison of our results for different tasks and environments with those of previous studies revealed that our proposed method improves the success rate by up to 29.3 points. We also conduct ablation studies using our proposed method. Learning task-related viewpoints from the multiple viewpoints dataset increases robustness to occlusion than using a uniquely defined viewpoint. Focusing on task-related regions contributes to up to a 33.3-point improvement in the success rate against domain shift.

cs.RO↗

arXiv2024

CLIP-Loc: Multi-modal Landmark Association for Global Localization in Object-based Maps

This paper describes a multi-modal data association method for global localization using object-based maps and camera images. In global localization, or relocalization, using object-based maps, existing methods typically resort to matching all possible combinations of detected objects and landmarks with the same object category, followed by inlier extraction using RANSAC or brute-force search. This approach becomes infeasible as the number of landmarks increases due to the exponential growth of correspondence candidates. In this paper, we propose labeling landmarks with natural language descriptions and extracting correspondences based on conceptual similarity with image observations using a Vision Language Model (VLM). By leveraging detailed text information, our approach efficiently extracts correspondences compared to methods using only object categories. Through experiments, we demonstrate that the proposed method enables more accurate global localization with fewer iterations compared to baseline methods, exhibiting its efficiency.

cs.CV↗