arXiv subjects

Rintaro Otsubo

Publications and source records attributed to Rintaro Otsubo.

4 recordsLinked to original sources

arXiv2026

AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models

Vision-Language Models (VLMs) have demonstrated immense promise in Spatio-Temporal Video Grounding (STVG). However, current evaluation protocols are largely confined to zero-shot assessments on general, daily-life benchmarks. This creates a critical disconnect from real-world applications in specialized fields, where models inevitably encounter rare visual concepts and complex spatio-temporal dynamics. Since exhaustive pre-training across infinite data distributions is infeasible, the ability to adapt to novel domains is essential. To bridge this gap, we introduce AnyGroundBench, a domain-adaptation benchmark designed to shift the STVG evaluation paradigm from static zero-shot testing to rigorous domain adaptation. Targeting five specialized domains (animal, industry, sports, surgery, and public security), AnyGroundBench pairs newly captured videos such as expert-annotated mouse behaviors with established datasets, unifying them through dense, high-fidelity spatio-temporal annotations. Crucially, the benchmark provides dedicated training subsets to systematically measure domain adaptability. We extensively evaluate 15 state-of-the-art VLMs, assessing their zero-shot generalization and In-Context Learning (ICL) capabilities under practical computational constraints. Ultimately, our findings reveal that current models fail in both zero-shot and ICL-based adaptation when confronted with specialized domains, exposing critical flaws in spatio-temporal reasoning that future research must address.

cs.CV↗

arXiv2025

Hand Held Multi-Object Tracking Dataset in American Football

Multi-Object Tracking (MOT) plays a critical role in analyzing player behavior from videos, enabling performance evaluation. Current MOT methods are often evaluated using publicly available datasets. However, most of these focus on everyday scenarios such as pedestrian tracking or are tailored to specific sports, including soccer and basketball. Despite the inherent challenges of tracking players in American football, such as frequent occlusion and physical contact, no standardized dataset has been publicly available, making fair comparisons between methods difficult. To address this gap, we constructed the first dedicated detection and tracking dataset for the American football players and conducted a comparative evaluation of various detection and tracking methods. Our results demonstrate that accurate detection and tracking can be achieved even in crowded scenarios. Fine-tuning detection models improved performance over pre-trained models. Furthermore, when these fine-tuned detectors and re-identification models were integrated into tracking systems, we observed notable improvements in tracking accuracy compared to existing approaches. This work thus enables robust detection and tracking of American football players in challenging, high-density scenarios previously underserved by conventional methods.

cs.CV↗

arXiv2025

SoccerNet 2025 Challenges Results

The SoccerNet 2025 Challenges mark the fifth annual edition of the SoccerNet open benchmarking effort, dedicated to advancing computer vision research in football video understanding. This year's challenges span four vision-based tasks: (1) Team Ball Action Spotting, focused on detecting ball-related actions in football broadcasts and assigning actions to teams; (2) Monocular Depth Estimation, targeting the recovery of scene geometry from single-camera broadcast clips through relative depth estimation for each pixel; (3) Multi-View Foul Recognition, requiring the analysis of multiple synchronized camera views to classify fouls and their severity; and (4) Game State Reconstruction, aimed at localizing and identifying all players from a broadcast video to reconstruct the game state on a 2D top-view of the field. Across all tasks, participants were provided with large-scale annotated datasets, unified evaluation protocols, and strong baselines as starting points. This report presents the results of each challenge, highlights the top-performing solutions, and provides insights into the progress made by the community. The SoccerNet Challenges continue to serve as a driving force for reproducible, open research at the intersection of computer vision, artificial intelligence, and sports. Detailed information about the tasks, challenges, and leaderboards can be found at https://www.soccer-net.org, with baselines and development kits available at https://github.com/SoccerNet.

cs.CV↗

arXiv2024

Advanced POD-Based Performance Evaluation of Classifiers Applied to Human Driver Lane Changing Prediction

Machine learning (ML) classifiers serve as essential tools facilitating classification and prediction across various domains. The performance of these algorithms should be known to ensure their reliable application. In certain fields, receiver operating characteristic and precision-recall curves are frequently employed to assess machine learning algorithms without accounting for the impact of process parameters. However, it may be essential to evaluate the performance of these algorithms in relation to such parameters. As a performance evaluation metric capable of considering the effects of process parameters, this paper uses a modified probability of detection (POD) approach to assess the reliability of ML-based algorithms. As an example, the POD-based approach is employed to assess ML models used for predicting the lane changing behavior of a vehicle driver. The time remaining to the predicted (and therefore unknown) lane changing event is considered as process parameter. The hit/miss approach to POD is taken here and modified by considering the probability of lane changing derived from ML algorithms at each time step, and obtaining the final result of the analysis accordingly. This improves the reliability of results compared to the standard hit/miss approach, which considers the outcome of the classifiers as either 0 or 1, while also simplifying evaluation compared to the â versus a approach. Performance evaluation results of the proposed approach are compared with those obtained with the standard hit/miss approach and a pre-developed â versus a approach to validate the effectiveness of the proposed method. The comparison shows that this method provides an averaging conservative behavior with the advantage of enhancing the reliability of the hit/miss approach to POD while retaining its simplicity.

eess.SY↗