arXiv subjects

Toshihiko Nishimura

Publications and source records attributed to Toshihiko Nishimura.

6 recordsLinked to original sources

arXiv2026

Probing Association Instability with Track-State Perturbations for Clip-Level Active Learning in Query-Propagation Multi-Object Tracking

Training query-propagation end-to-end multi-object tracking (MOT) models requires dense bounding-box and identity annotations across video sequences, making dataset construction expensive. Clip-level active learning reduces this cost by selecting video clips for annotation, but prior acquisition criteria based on output-level temporal uncertainty may miss clips whose informativeness comes from association instability in propagated track states. We propose QPID (Query-Propagation Instability and Diversity), a clip acquisition method for query-propagation MOT that targets association instability in propagated track states. QPID estimates this instability by applying two-sided perturbations to internal track states and measuring prediction differences from a clean reference branch. The key idea is that, in stable clips, each propagated track should continue to follow the same target under small perturbations, whereas in ambiguous clips, small changes in the track state can alter which target the track follows, leading to changes in localization or confidence. QPID measures these perturbation-induced prediction differences with two metrics: Localization Drift and Entropy-Weighted Confidence Discrepancy. These metrics are aggregated into a clip-level association-instability score. To avoid redundant uncertainty-only selection, QPID selects a representative annotation batch from high-instability clips using Uncertainty-Weighted Visual Coverage with track-level visual prototypes. Experiments on DanceTrack and SportsMOT with MeMOTR and SambaMOTR show that QPID achieves strong performance compared with active learning baselines under the same annotation budget.

cs.CV↗

arXiv2026

Clip-level Uncertainty and Temporal-aware Active Learning for End-to-End Multi-Object Tracking

Multi-Object Tracking (MOT) in dynamic environments relies on robust temporal reasoning to maintain consistent object identities over time. Transformer-based end-to-end MOT models achieve strong performance by explicitly modeling temporal dependencies, yet training them requires extensive bounding-box and identity annotations. Given the high labeling cost and strong redundancy in videos, Active Learning (AL) is an effective approach to improve annotation efficiency. However, existing AL methods for MOT primarily operate at the frame level, which is structurally misaligned with modern end-to-end trackers whose inference and training rely on multi-frame clips. To bridge this gap, we formulate clip-level active learning and propose Clip-level Uncertainty and Temporal-aware Active Learning (CUTAL). In contrast to frame-based approaches, CUTAL scores each clip using uncertainty metrics derived from multi-frame predictions to capture inter-frame correspondence ambiguities, while enforcing temporal diversity to select an informative and non-redundant subset. Experiments show that CUTAL achieves stronger overall performance than baselines at the same label budgets across MeMOTR and SambaMOTR. Notably, CUTAL achieves performance comparable to full supervision for MeMOTR on both datasets using only 50% of the labeled training data.

cs.CV↗

arXiv2026

Leveraging 2D-VLM for Label-Free 3D Segmentation in Large-Scale Outdoor Scene Understanding

This paper presents a novel 3D semantic segmentation method for large-scale point cloud data that does not require annotated 3D training data or paired RGB images. The proposed approach projects 3D point clouds onto 2D images using virtual cameras and performs semantic segmentation via a foundation 2D model guided by natural language prompts. 3D segmentation is achieved by aggregating predictions from multiple viewpoints through weighted voting. Our method outperforms existing training-free approaches and achieves segmentation accuracy comparable to supervised methods. Moreover, it supports open-vocabulary recognition, enabling users to detect objects using arbitrary text queries, thus overcoming the limitations of traditional supervised approaches.

cs.CV↗

arXiv2025

Hamiltonian Monte Carlo-Based Near-Optimal MIMO Signal Detection

Multiple-input multiple-output (MIMO) technology is essential for the optimal functioning of next-generation wireless networks; however, enhancing its signal-detection performance for improved spectral efficiency is challenging. Here, we propose an approach that transforms the discrete MIMO detection problem into a continuous problem while leveraging the efficient Hamiltonian Monte Carlo algorithm. For this continuous framework, we employ a mixture of t-distributions as the prior distribution. To improve the performance in the coded case further, we treat the likelihood's temperature parameter as a random variable and address its optimization. This treatment leads to the adoption of a horseshoe density for the likelihood. Theoretical analysis and extensive simulations demonstrate that our method achieves near-optimal detection performance while maintaining polynomial computational complexity. This MIMO detection technique can accelerate the development of 6G mobile communication systems.

cs.NI↗

arXiv2024

Memory-Efficient Point Cloud Registration via Overlapping Region Sampling

Recent advances in deep learning have improved 3D point cloud registration but increased graphics processing unit (GPU) memory usage, often requiring preliminary sampling that reduces accuracy. We propose an overlapping region sampling method to reduce memory usage while maintaining accuracy. Our approach estimates the overlapping region and intensively samples from it, using a k-nearest-neighbor (kNN) based point compression mechanism with multi layer perceptron (MLP) and transformer architectures. Evaluations on 3DMatch and 3DLoMatch datasets show our method outperforms other sampling methods in registration recall, especially at lower GPU memory levels. For 3DMatch, we achieve 94% recall with 33% reduced memory usage, with greater advantages in 3DLoMatch. Our method enables efficient large-scale point cloud registration in resource-constrained environments, maintaining high accuracy while significantly reducing memory requirements.

cs.CV↗

arXiv2024

Near-optimal stochastic MIMO signal detection with a mixture of t-distribution prior

Multiple-input multiple-output (MIMO) systems will play a crucial role in future wireless communication, but improving their signal detection performance to increase transmission efficiency remains a challenge. To address this issue, we propose extending the discrete signal detection problem in MIMO systems to a continuous one and applying the Hamiltonian Monte Carlo method, an efficient Markov chain Monte Carlo algorithm. In our previous studies, we have used a mixture of normal distributions for the prior distribution. In this study, we propose using a mixture of t-distributions, which further improves detection performance. Based on our theoretical analysis and computer simulations, the proposed method can achieve near-optimal signal detection with polynomial computational complexity. This high-performance and practical MIMO signal detection could contribute to the development of the 6th-generation mobile network.

cs.NI↗