The SoccerNet 2026 Challenges constitute the sixth annual edition of the SoccerNet open benchmarking effort, dedicated to advancing computer vision research in sports video understanding. This year's challenges span five vision-based tasks: (1) Ball Action Anticipation, predicting the timing and class of ball-related actions within a short future window from a preceding observation window; (2) Player-Centric Ball Action Spotting, temporally localizing and classifying ball-related actions while assigning each action to the acting player through team affiliation and jersey number; (3) Novel View Synthesis, rendering images from unobserved camera poses in multi-view football scenes; (4) Spiideo SoccerNet Synloc, localizing athletes in real-world pitch coordinates from a single calibrated static-camera image; and (5) Visual Question Answering, answering multiple-choice questions about football broadcasts across text, image, and video inputs. For each task, participants were provided with annotated data, a unified evaluation protocol, and a public baseline. This edition saw broad participation, with 427 teams submitting 1,129 entries across the five tasks and 28 teams contributing reviewed technical reports. This paper describes each task and its evaluation protocol, presents the challenge leaderboards, and summarizes the leading submissions, with the aim of documenting the current state of each task as measured on held-out challenge data.
A. Cioppa, Silvio Giancola, Haakan Ardo et al.· 0 citations
Unstructured environments challenge unmanned ground vehicle (UGV) navigation with complex terrain and open-set obstacles. Existing inertial-aided navigation using external odometry suffers from localization errors in rugged off-road conditions, while sparse LiDAR point clouds degrade traversability prediction. The purpose of this study is to address these limitations by developing a robust autonomous navigation framework that integrates SLAM-assisted normal distributions transform (SANDT) and divergence-guided temporal point cloud fusion.
First, the authors replace conventional vehicle odometry with inertial data maintained by a LiDAR-based SLAM method, supplying a continuous and stable coarse guess for NDT registration to improve localization. Second, voxel-wise NDT representations of adjacent point clouds are computed; key historical frames are selected via Jensen-Shannon divergence and fused to densify the current point cloud and improve traversability estimation. Finally, the authors integrate these components into an autonomous navigation framework and validate it in real-world scenarios.
Experiments demonstrate that the authors’ framework achieves accurate localization and seamless indoor-to-outdoor navigation, outperforming baseline methods in traversability prediction, navigation success rate and obstacle avoidance.
This paper presents a robust autonomous navigation framework for unstructured environments. SANDT enables cross-scene navigation in complex terrains, and global divergence-based temporal fusion pioneers LiDAR-based traversability estimation. Further details on localization, traversability prediction and real-world navigation are provided in the supplementary video.
Yuenan Zhao, Ziming Zhang, Ruifeng Wang et al.· Robotic Intelligence and Aut...· 0 citations