Calvin Yeung

Yuchen Hou, Yongshan Chen, Zhuowen Zou et al.

Federated reinforcement learning enables decentralized agents to collaboratively improve policies or value estimates without exchanging raw trajectories. However, FedAvg-style parameter averaging is not function-space consistent: when clients use heterogeneous encoders or even identical nonlinear networks, averaged parameters need not correspond to the weighted average of client value functions in any common function space. We propose FedQHD, a federated Q-learning method using hyperdimensional (random-feature) state encoders with a linear readout, so that Q-functions are nonlinear in state yet linear in trainable parameters. This linear structure enables closed-form aggregation. With a shared encoder, the function-space consensus update coincides exactly with weighted averaging of local readout matrices. With heterogeneous encoders, the server constructs a global teacher by averaging client Q-values on a shared anchor-state set, and each client compiles this teacher into its local representation via a single ridge projection. We formalize the federation gap -- the error incurred when compiling a federated teacher into a heterogeneous client representation -- relative to a client-specific oracle projection. We show that this gap decomposes into subspace misalignment, anchor-set conditioning, and regularization bias. We further identify the anchor-to-dimension ratio $m \geq D_i$ as the well-conditioned regime in which the gap reduces to a multiple of the encoder heterogeneity floor. On four continuous-state, discrete-action control benchmarks, FedQHD matches or outperforms FedAvg-style baselines and distillation-based alternatives while requiring substantially less computation, and the empirical dependence of the federation gap on encoder dimension matches our theoretical analysis.

Calvin Yeung, Prathyush Poduval, Ali Zakeri et al.

Text-to-image diffusion models generate images through an iterative denoising process, so internal neural layers produce trajectories of activations rather than single static representations. Sparse autoencoders (SAEs) have recently been used to decompose diffusion activations into interpretable feature directions, but most approaches analyze activations at individual timesteps or condition on time rather than learning directly from full activation trajectories. In this work, we introduce residualized temporal SAEs for diffusion activation trajectories. We collect activations across denoising time, fit linear predictors between neighboring timesteps, and represent each trajectory using an initial activation together with residual components not explained by these linear dynamics. Training an SAE on this residualized representation encourages sparse latents to capture structure beyond what is linearly predictable. The residualized decoder directions can be mapped back into activation space, allowing each latent to be analyzed as a feature trajectory over denoising time. Through reconstruction and ablation studies, spatiotemporal feature analysis, and qualitative steering experiments on Stable Diffusion~1.5, we show that residualized temporal SAEs provide a useful framework for studying temporally structured diffusion activations.

Prathyush Poduval, Calvin Yeung, Neel Desai et al.

Sparse autoencoders are usually trained one layer at a time, even though transformer residual stream activations are strongly coupled across depth. This creates a practical problem for multi-layer interventions: different layerwise dictionaries can spend capacity representing the same carried-forward information, and replacing several layers at once can produce interactions that are not predicted by single-layer behavior. We introduce Residualized Sparse Autoencoders (ReSAEs), which fit an affine map between selected layers and train each later-layer SAE on the unexplained residual rather than on the full activation. Reconstructions are mapped back into the original activation space through the fitted affine chain, so ReSAEs can be evaluated with the same intervention protocols as ordinary SAEs. On Pythia-1.4B and Gemma-2-9B, residualization reduces decoder redundancy and improves sparse probing and targeted perturbation in most tested settings. Despite reconstructing less of the raw activation variance, ReSAEs recover more transformer cross entropy under multi-layer replacement. This gain is clearest under teacher-forcing and at sufficient sparsity online, indicating that ReSAEs preserve the components of the activation most relevant to the model's downstream computation. These results suggest that removing linearly predictable cross-layer structure is a useful default for multi-layer SAE interventions.

Anthony Cioppa, Silvio Giancola, Vladimir Somers et al.

The SoccerNet 2024 challenges represent the fourth annual video understanding challenges organized by the SoccerNet team. These challenges aim to advance research across multiple themes in football, including broadcast video understanding, field understanding, and player understanding. This year, the challenges encompass four vision-based tasks. (1) Ball Action Spotting, focusing on precisely localizing when and which soccer actions related to the ball occur, (2) Dense Video Captioning, focusing on describing the broadcast with natural language and anchored timestamps, (3) Multi-View Foul Recognition, a novel task focusing on analyzing multiple viewpoints of a potential foul incident to classify whether a foul occurred and assess its severity, (4) Game State Reconstruction, another novel task focusing on reconstructing the game state from broadcast videos onto a 2D top-view map of the field. 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.

Li Yin, Qin Haobin, Tomohiro Suzuki et al.

Multi-view multi-person 3D pose estimation in team sports scenarios remains challenging due to player occlusions, appearance similarity caused by team uniforms, and the scarcity of annotated multi-view data, all of which limit the effectiveness and generalization capability of learning-based methods. In contrast, the performance of training-free approaches is inherently constrained by the accuracy of 2D keypoint detection and the robustness of cross-view association. To address these challenges, we propose Mesh-Aware Epipolar Matching (MAEM), a training-free framework for multi-view multi-person 3D pose estimation. Our method employs a monocular 3D human mesh recovery model as the frontend and introduces a two-stage epipolar matching strategy based on the recovered mesh outputs. Specifically, the proposed framework combines disjoint-set-union-based clustering with per-joint triangulation to achieve robust cross-view association and accurate 3D pose reconstruction. Experiments on two public multi-view basketball datasets demonstrate that MAEM consistently outperforms existing training-free association baselines while achieving competitive RGB-only performance in both indoor and outdoor basketball scenarios. MAEM achieves MPJPE/PA-MPJPE scores of 59.8/40.7 mm on SportCenter EPFL and 74.0/51.8 mm on Human-M3 Basketball, highlighting the effectiveness of dense mesh geometry for cross-view association without requiring target-domain training or fine-tuning.

Calvin Yeung, Rory Bunker, Rikuhei Umemoto et al.

Machine learning models have become increasingly popular for predicting the results of soccer matches, however, the lack of publicly-available benchmark datasets has made model evaluation challenging. The 2023 Soccer Prediction Challenge required the prediction of match results first in terms of the exact goals scored by each team, and second, in terms of the probabilities for a win, draw, and loss. The original training set of matches and features, which was provided for the competition, was augmented with additional matches that were played between 4 April and 13 April 2023, representing the period after which the training set ended, but prior to the first matches that were to be predicted (upon which the performance was evaluated). A CatBoost model was employed using pi-ratings as the features, which were initially identified as the optimal choice for calculating the win/draw/loss probabilities. Notably, deep learning models have frequently been disregarded in this particular task. Therefore, in this study, we aimed to assess the performance of a deep learning model and determine the optimal feature set for a gradient-boosted tree model. The model was trained using the most recent five years of data, and three training and validation sets were used in a hyperparameter grid search. The results from the validation sets show that our model had strong performance and stability compared to previously published models from the 2017 Soccer Prediction Challenge for win/draw/loss prediction.

Calvin Yeung, Kenjiro Ide, Keisuke Fujii

Image understanding is a foundational task in computer vision, with recent applications emerging in soccer posture analysis. However, existing publicly available datasets lack comprehensive information, notably in the form of posture sequences and 2D pose annotations. Moreover, current analysis models often rely on interpretable linear models (e.g., PCA and regression), limiting their capacity to capture non-linear spatiotemporal relationships in complex and diverse scenarios. To address these gaps, we introduce the 3D Shot Posture (3DSP) dataset in soccer broadcast videos, which represents the most extensive sports image dataset with 2D pose annotations to our knowledge. Additionally, we present the 3DSP-GRAE (Graph Recurrent AutoEncoder) model, a non-linear approach for embedding pose sequences. Furthermore, we propose AutoSoccerPose, a pipeline aimed at semi-automating 2D and 3D pose estimation and posture analysis. While achieving full automation proved challenging, we provide a foundational baseline, extending its utility beyond the scope of annotated data. We validate AutoSoccerPose on SoccerNet and 3DSP datasets, and present posture analysis results based on 3DSP. The dataset, code, and models are available at: https://github.com/calvinyeungck/3D-Shot-Posture-Dataset.

Calvin Yeung, Tomohiro Suzuki, Ryota Tanaka et al.

Human pose estimation is a critical task in computer vision and sports biomechanics, with applications spanning sports science, rehabilitation, and biomechanical research. While significant progress has been made in monocular 3D pose estimation, current datasets often fail to capture the complex, high-acceleration movements typical of competitive sports. In this work, we introduce AthletePose3D, a novel dataset designed to address this gap. AthletePose3D includes 12 types of sports motions across various disciplines, with approximately 1.3 million frames and 165 thousand individual postures, specifically capturing high-speed, high-acceleration athletic movements. We evaluate state-of-the-art (SOTA) monocular 2D and 3D pose estimation models on the dataset, revealing that models trained on conventional datasets perform poorly on athletic motions. However, fine-tuning these models on AthletePose3D notably reduces the SOTA model mean per joint position error (MPJPE) from 214mm to 65mm-a reduction of over 69%. We also validate the kinematic accuracy of monocular pose estimations through waveform analysis, highlighting strong correlations in joint angle estimations but limitations in velocity estimation. Our work provides a comprehensive evaluation of monocular pose estimation models in the context of sports, contributing valuable insights for advancing monocular pose estimation techniques in high-performance sports environments. The dataset, code, and model checkpoints are available at: https://github.com/calvinyeungck/AthletePose3D

Calvin Yeung, Ali Zakeri, Zhuowen Zou et al.

Many visual scenes can be described as compositions of latent factors. Effective recognition, reasoning, and editing often require not only forming such compositional representations, but also solving the decomposition problem. One popular choice for constructing these representations is through the binding operation. Resonator networks, which can be understood as coupled Hopfield networks, were proposed as a way to perform decomposition on such bound representations. Recent works have shown notable similarities between Hopfield networks and diffusion models. Motivated by these observations, we introduce a framework for semantic decomposition using coupled inference in diffusion models. Our method frames semantic decomposition as an inverse problem and couples the diffusion processes using a reconstruction-driven guidance term that encourages the composition of factor estimates to match the bound vector. We also introduce a novel iterative sampling scheme that improves the performance of our model. Finally, we show that attention-based resonator networks are a special case of our framework. Empirically, we demonstrate that our coupled inference framework outperforms resonator networks across a range of synthetic semantic decomposition tasks.

William Youngwoo Chung, Calvin Yeung, Hansen Jin Lillemark et al.

A key challenge in artificial intelligence and neuroscience is understanding how neural systems learn representations that capture the underlying dynamics of the world. Most world models represent the transition function with unstructured neural networks, limiting interpretability, sample efficiency, and generalization to unseen states or action compositions. We address these issues with a generalizable world model grounded in Vector Symbolic Architecture (VSA) principles as geometric priors. Our approach utilizes learnable Fourier Holographic Reduced Representation (FHRR) encoders to map states and actions into a high dimensional complex vector space with learned group structure and models transitions with element-wise complex multiplication. We formalize the framework's group theoretic foundation and show how training such structured representations to be approximately invariant enables strong multi-step composition directly in latent space and generalization performances over various experiments. On a discrete grid world environment, our model achieves 87.5% zero shot accuracy to unseen state-action pairs, obtains 53.6% higher accuracy on 20-timestep horizon rollouts, and demonstrates 4x higher robustness to noise relative to an MLP baseline. These results highlight how training to have latent group structure yields generalizable, data-efficient, and interpretable world models, providing a principled pathway toward structured models for real-world planning and reasoning.

Calvin Yeung, Kenjiro Ide, Taiga Someya et al.

Sports analytics has become both more professional and sophisticated, driven by the growing availability of detailed performance data. This progress enables applications such as match outcome prediction, player scouting, and tactical analysis. In soccer, the effective utilization of event and tracking data is fundamental for capturing and analyzing the dynamics of the game. However, there are two primary challenges: the limited availability of event data, primarily restricted to top-tier teams and leagues, and the scarcity and high cost of tracking data, which complicates its integration with event data for comprehensive analysis. Here we propose OpenSTARLab, an open-source framework designed to democratize spatio-temporal agent data analysis in sports by addressing these key challenges. OpenSTARLab includes the Pre-processing Package that standardizes event and tracking data through Unified and Integrated Event Data and State-Action-Reward formats, the Event Modeling Package that implements deep learning-based event prediction, alongside the RLearn Package for reinforcement learning tasks. These technical components facilitate the handling of diverse data sources and support advanced analytical tasks, thereby enhancing the overall functionality and usability of the framework. To assess OpenSTARLab's effectiveness, we conducted several experimental evaluations. These demonstrate the superior performance of the specific event prediction model in terms of action and time prediction accuracies and maintained its robust event simulation performance. Furthermore, reinforcement learning experiments reveal a trade-off between action accuracy and temporal difference loss and show comprehensive visualization. Overall, OpenSTARLab serves as a robust platform for researchers and practitioners, enhancing innovation and collaboration in the field of soccer data analytics.

Kazuhiro Yamada, Li Yin, Qingrui Hu et al.

Multi-object tracking, player identification, and pose estimation are fundamental components of sports analytics, essential for analyzing player movements, performance, and tactical strategies. However, existing datasets and methodologies primarily target mainstream team sports such as soccer and conventional 5-on-5 basketball, often overlooking scenarios involving fixed-camera setups commonly used at amateur levels, less mainstream sports, or datasets that explicitly incorporate pose annotations. In this paper, we propose the TrackID3x3 dataset, the first publicly available comprehensive dataset specifically designed for multi-player tracking, player identification, and pose estimation in 3x3 basketball scenarios. The dataset comprises three distinct subsets (Indoor fixed-camera, Outdoor fixed-camera, and Drone camera footage), capturing diverse full-court camera perspectives and environments. We also introduce the Track-ID task, a simplified variant of the game state reconstruction task that excludes field detection and focuses exclusively on fixed-camera scenarios. To evaluate performance, we propose a baseline algorithm called Track-ID algorithm, tailored to assess tracking and identification quality. Furthermore, our benchmark experiments, utilizing recent multi-object tracking algorithms (e.g., BoT-SORT-ReID) and top-down pose estimation methods (HRNet, RTMPose, and SwinPose), demonstrate robust results and highlight remaining challenges. Our dataset and evaluation benchmarks provide a solid foundation for advancing automated analytics in 3x3 basketball. Dataset and code will be available at https://github.com/open-starlab/TrackID3x3.

Yezi Liu, William Youngwoo Chung, Hanning Chen et al.

Recent advances in large language models (LLMs) have enabled strong reasoning over both structured and unstructured knowledge. When grounded on knowledge graphs (KGs), however, prevailing pipelines rely on heavy neural encoders to embed and score symbolic paths or on repeated LLM calls to rank candidates, leading to high latency, GPU cost, and opaque decisions that hinder faithful, scalable deployment. We propose PathHD, a lightweight and encoder-free KG reasoning framework that replaces neural path scoring with hyperdimensional computing (HDC) and uses only a single LLM call per query. PathHD encodes relation paths into block-diagonal GHRR hypervectors, ranks candidates with blockwise cosine similarity and Top-K pruning, and then performs a one-shot LLM adjudication to produce the final answer together with cited supporting paths. Technically, PathHD is built on three ingredients: (i) an order-aware, non-commutative binding operator for path composition, (ii) a calibrated similarity for robust hypervector-based retrieval, and (iii) a one-shot adjudication step that preserves interpretability while eliminating per-path LLM scoring. On WebQSP, CWQ, and the GrailQA split, PathHD (i) attains comparable or better Hits@1 than strong neural baselines while using one LLM call per query; (ii) reduces end-to-end latency by $40-60\%$ and GPU memory by $3-5\times$ thanks to encoder-free retrieval; and (iii) delivers faithful, path-grounded rationales that improve error diagnosis and controllability. These results indicate that carefully designed HDC representations provide a practical substrate for efficient KG-LLM reasoning, offering a favorable accuracy-efficiency-interpretability trade-off.

Tomohiro Suzuki, Ryota Tanaka, Calvin Yeung et al.

Monocular 3D pose estimation is a promising, flexible alternative to costly motion capture systems for sports analysis. However, its practical application is hindered by two factors: a lack of realistic sports datasets and unclear reliability for sports tasks. To address these challenges, we introduce the AthleticsPose dataset, a new public dataset featuring ``real'' motions captured from 23 athletes performing various athletics events on an athletic field. Using this dataset, we trained a representative 3D pose estimation model and performed a comprehensive evaluation. Our results show that the model trained on AthleticsPose significantly outperforms a baseline model trained on an imitated sports motion dataset, reducing MPJPE by approximately 75 %. These results show the importance of training on authentic sports motion data, as models based on imitated motions do not effectively transfer to real-world motions. Further analysis reveals that estimation accuracy is sensitive to camera view and subject scale. In case studies of kinematic indicators, the model demonstrated the potential to capture individual differences in knee angles but struggled with higher-speed metrics, such as knee-drive velocity, due to prediction biases. This work provides the research community with a valuable dataset and clarifies the potential and practical limitations of using monocular 3D pose estimation for sports motion analysis. Our dataset, code, and checkpoints are available at https://github.com/SZucchini/AthleticsPose.

Zhuoer Yin, Calvin Yeung, Tomohiro Suzuki et al.

Recent transformer based approaches have demonstrated impressive performance in solving real-world 3D human pose estimation problems. Albeit these approaches achieve fruitful results on benchmark datasets, they tend to fall short of sports scenarios where human movements are more complicated than daily life actions, as being hindered by motion blur, occlusions, and domain shifts. Moreover, due to the fact that critical motions in a sports game often finish in moments of time (e.g., shooting), the ability to focus on momentary actions is becoming a crucial factor in sports analysis, where current methods appear to struggle with instantaneous scenarios. To overcome these limitations, we introduce KASportsFormer, a novel transformer based 3D pose estimation framework for sports that incorporates a kinematic anatomy-informed feature representation and integration module. In which the inherent kinematic motion information is extracted with the Bone Extractor (BoneExt) and Limb Fuser (LimbFus) modules and encoded in a multimodal manner. This improved the capability of comprehending sports poses in short videos. We evaluate our method through two representative sports scene datasets: SportsPose and WorldPose. Experimental results show that our proposed method achieves state-of-the-art results with MPJPE errors of 58.0mm and 34.3mm, respectively. Our code and models are available at: https://github.com/jw0r1n/KASportsFormer

Jiale Fang, Calvin Yeung, Keisuke Fujii

Recent advances in computer vision have made significant progress in tracking and pose estimation of sports players. However, there have been fewer studies on behavior prediction with pose estimation in sports, in particular, the prediction of soccer fouls is challenging because of the smaller image size of each player and of difficulty in the usage of e.g., the ball and pose information. In our research, we introduce an innovative deep learning approach for anticipating soccer fouls. This method integrates video data, bounding box positions, image details, and pose information by curating a novel soccer foul dataset. Our model utilizes a combination of convolutional and recurrent neural networks (CNNs and RNNs) to effectively merge information from these four modalities. The experimental results show that our full model outperformed the ablated models, and all of the RNN modules, bounding box position and image, and estimated pose were useful for the foul prediction. Our findings have important implications for a deeper understanding of foul play in soccer and provide a valuable reference for future research and practice in this area.

Calvin Yeung, Zhuowen Zou, Mohsen Imani

Deep learning has achieved remarkable success in recent years. Central to its success is its ability to learn representations that preserve task-relevant structure. However, massive energy, compute, and data costs are required to learn general representations. This paper explores Hyperdimensional Computing (HDC), a computationally and data-efficient brain-inspired alternative. HDC acts as a bridge between connectionist and symbolic approaches to artificial intelligence (AI), allowing explicit specification of representational structure as in symbolic approaches while retaining the flexibility of connectionist approaches. However, HDC's simplicity poses challenges for encoding complex compositional structures, especially in its binding operation. To address this, we propose Generalized Holographic Reduced Representations (GHRR), an extension of Fourier Holographic Reduced Representations (FHRR), a specific HDC implementation. GHRR introduces a flexible, non-commutative binding operation, enabling improved encoding of complex data structures while preserving HDC's desirable properties of robustness and transparency. In this work, we introduce the GHRR framework, prove its theoretical properties and its adherence to HDC properties, explore its kernel and binding characteristics, and perform empirical experiments showcasing its flexible non-commutativity, enhanced decoding accuracy for compositional structures, and improved memorization capacity compared to FHRR.

Calvin Yeung, Jeff Yu, King Chau Cheung et al.

Automated grading has become an essential tool in education technology due to its ability to efficiently assess large volumes of student work, provide consistent and unbiased evaluations, and deliver immediate feedback to enhance learning. However, current systems face significant limitations, including the need for large datasets in few-shot learning methods, a lack of personalized and actionable feedback, and an overemphasis on benchmark performance rather than student experience. To address these challenges, we propose a Zero-Shot Large Language Model (LLM)-Based Automated Assignment Grading (AAG) system. This framework leverages prompt engineering to evaluate both computational and explanatory student responses without requiring additional training or fine-tuning. The AAG system delivers tailored feedback that highlights individual strengths and areas for improvement, thereby enhancing student learning outcomes. Our study demonstrates the system's effectiveness through comprehensive evaluations, including survey responses from higher education students that indicate significant improvements in motivation, understanding, and preparedness compared to traditional grading methods. The results validate the AAG system's potential to transform educational assessment by prioritizing learning experiences and providing scalable, high-quality feedback.

Rory Bunker, Calvin Yeung, Keisuke Fujii

Machine learning has become a common approach to predicting the outcomes of soccer matches, and the body of literature in this domain has grown substantially in the past decade and a half. This chapter discusses available datasets, the types of models and features, and ways of evaluating model performance in this application domain. The aim of this chapter is to give a broad overview of the current state and potential future developments in machine learning for soccer match results prediction, as a resource for those interested in conducting future studies in the area. Our main findings are that while gradient-boosted tree models such as CatBoost, applied to soccer-specific ratings such as pi-ratings, are currently the best-performing models on datasets containing only goals as the match features, there needs to be a more thorough comparison of the performance of deep learning models and Random Forest on a range of datasets with different types of features. Furthermore, new rating systems using both player- and team-level information and incorporating additional information from, e.g., spatiotemporal tracking and event data, could be investigated further. Finally, the interpretability of match result prediction models needs to be enhanced for them to be more useful for team management.

Li Yin, Calvin Yeung, Qingrui Hu et al.

Multi-object tracking (MOT) is crucial for various multi-agent analyses such as evaluating team sports tactics and player movements and performance. While pedestrian tracking has advanced with Tracking-by-Detection MOT, team sports like basketball pose unique challenges. These challenges include players' unpredictable movements, frequent close interactions, and visual similarities that complicate pose labeling and lead to significant occlusions, frequent ID switches, and high manual annotation costs. To address these challenges, we propose a novel pose-based virtual marker (VM) MOT method for team sports, named Sports-vmTracking. This method builds on the vmTracking approach developed for multi-animal tracking with active learning. First, we constructed a 3x3 basketball pose dataset for VMs and applied active learning to enhance model performance in generating VMs. Then, we overlaid the VMs on video to identify players, extract their poses with unique IDs, and convert these into bounding boxes for comparison with automated MOT methods. Using our 3x3 basketball dataset, we demonstrated that our VM configuration has been highly effective, and reduced the need for manual corrections and labeling during pose model training while maintaining high accuracy. Our approach achieved an average HOTA score of 72.3%, over 10 points higher than other state-of-the-art methods without VM, and resulted in 0 ID switches. Beyond improving performance in handling occlusions and minimizing ID switches, our framework could substantially increase the time and cost efficiency compared to traditional manual annotation.

Haobin Qin, Calvin Yeung, Rikuhei Umemoto et al.

In soccer video analysis, player detection is essential for identifying key events and reconstructing tactical positions. The presence of numerous players and frequent occlusions, combined with copyright restrictions, severely restricts the availability of datasets, leaving limited options such as SoccerNet-Tracking and SportsMOT. These datasets suffer from a lack of diversity, which hinders algorithms from adapting effectively to varied soccer video contexts. To address these challenges, we developed SoccerSynth-Detection, the first synthetic dataset designed for the detection of synthetic soccer players. It includes a broad range of random lighting and textures, as well as simulated camera motion blur. We validated its efficacy using the object detection model (Yolov8n) against real-world datasets (SoccerNet-Tracking and SportsMoT). In transfer tests, it matched the performance of real datasets and significantly outperformed them in images with motion blur; in pre-training tests, it demonstrated its efficacy as a pre-training dataset, significantly enhancing the algorithm's overall performance. Our work demonstrates the potential of synthetic datasets to replace real datasets for algorithm training in the field of soccer video analysis.

Calvin Yeung, Prathyush Poduval, Mohsen Imani

Vector Symbolic Architectures (VSAs) have emerged as a novel framework for enabling interpretable machine learning algorithms equipped with the ability to reason and explain their decision processes. The basic idea is to represent discrete information through high dimensional random vectors. Complex data structures can be built up with operations over vectors such as the "binding" operation involving element-wise vector multiplication, which associates data together. The reverse task of decomposing the associated elements is a combinatorially hard task, with an exponentially large search space. The main algorithm for performing this search is the resonator network, inspired by Hopfield network-based memory search operations. In this work, we introduce a new variant of the resonator network, based on self-attention based update rules in the iterative search problem. This update rule, based on the Hopfield network with log-sum-exp energy function and norm-bounded states, is shown to substantially improve the performance and rate of convergence. As a result, our algorithm enables a larger capacity for associative memory, enabling applications in many tasks like perception based pattern recognition, scene decomposition, and object reasoning. We substantiate our algorithm with a thorough evaluation and comparisons to baselines.

Silvio Giancola, Anthony Cioppa, Marc Gutiérrez-Pérez et al.

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.

Qingrui Hu, Atom Scott, Calvin Yeung et al.

Recent deep learning-based object detection approaches have led to significant progress in multi-object tracking (MOT) algorithms. The current MOT methods mainly focus on pedestrian or vehicle scenes, but basketball sports scenes are usually accompanied by three or more object occlusion problems with similar appearances and high-intensity complex motions, which we call complex multi-object occlusion (CMOO). Here, we propose an online and robust MOT approach, named Basketball-SORT, which focuses on the CMOO problems in basketball videos. To overcome the CMOO problem, instead of using the intersection-over-union-based (IoU-based) approach, we use the trajectories of neighboring frames based on the projected positions of the players. Our method designs the basketball game restriction (BGR) and reacquiring Long-Lost IDs (RLLI) based on the characteristics of basketball scenes, and we also solve the occlusion problem based on the player trajectories and appearance features. Experimental results show that our method achieves a Higher Order Tracking Accuracy (HOTA) score of 63.48$\%$ on the basketball fixed video dataset and outperforms other recent popular approaches. Overall, our approach solved the CMOO problem more effectively than recent MOT algorithms.