Sergio Escalera

Benjia Zhou, Zhigang Chen, Albert Clapés et al.

Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, involving the translation of visual-gestural language to text. Many previous methods employ an intermediate representation, i.e., gloss sequences, to facilitate SLT, thus transforming it into a two-stage task of sign language recognition (SLR) followed by sign language translation (SLT). However, the scarcity of gloss-annotated sign language data, combined with the information bottleneck in the mid-level gloss representation, has hindered the further development of the SLT task. To address this challenge, we propose a novel Gloss-Free SLT based on Visual-Language Pretraining (GFSLT-VLP), which improves SLT by inheriting language-oriented prior knowledge from pre-trained models, without any gloss annotation assistance. Our approach involves two stages: (i) integrating Contrastive Language-Image Pre-training (CLIP) with masked self-supervised learning to create pre-tasks that bridge the semantic gap between visual and textual representations and restore masked sentences, and (ii) constructing an end-to-end architecture with an encoder-decoder-like structure that inherits the parameters of the pre-trained Visual Encoder and Text Decoder from the first stage. The seamless combination of these novel designs forms a robust sign language representation and significantly improves gloss-free sign language translation. In particular, we have achieved unprecedented improvements in terms of BLEU-4 score on the PHOENIX14T dataset (>+5) and the CSL-Daily dataset (>+3) compared to state-of-the-art gloss-free SLT methods. Furthermore, our approach also achieves competitive results on the PHOENIX14T dataset when compared with most of the gloss-based methods. Our code is available at https://github.com/zhoubenjia/GFSLT-VLP.

Siyang Song, Micol Spitale, Cheng Luo et al.

The Multi-modal Multiple Appropriate Facial Reaction Generation Challenge (REACT2023) is the first competition event focused on evaluating multimedia processing and machine learning techniques for generating human-appropriate facial reactions in various dyadic interaction scenarios, with all participants competing strictly under the same conditions. The goal of the challenge is to provide the first benchmark test set for multi-modal information processing and to foster collaboration among the audio, visual, and audio-visual affective computing communities, to compare the relative merits of the approaches to automatic appropriate facial reaction generation under different spontaneous dyadic interaction conditions. This paper presents: (i) novelties, contributions and guidelines of the REACT2023 challenge; (ii) the dataset utilized in the challenge; and (iii) the performance of baseline systems on the two proposed sub-challenges: Offline Multiple Appropriate Facial Reaction Generation and Online Multiple Appropriate Facial Reaction Generation, respectively. The challenge baseline code is publicly available at \url{https://github.com/reactmultimodalchallenge/baseline_react2023}.

Alex Falcon, Swathikiran Sudhakaran, Giuseppe Serra et al.

Video retrieval using natural language queries has attracted increasing interest due to its relevance in real-world applications, from intelligent access in private media galleries to web-scale video search. Learning the cross-similarity of video and text in a joint embedding space is the dominant approach. To do so, a contrastive loss is usually employed because it organizes the embedding space by putting similar items close and dissimilar items far. This framework leads to competitive recall rates, as they solely focus on the rank of the groundtruth items. Yet, assessing the quality of the ranking list is of utmost importance when considering intelligent retrieval systems, since multiple items may share similar semantics, hence a high relevance. Moreover, the aforementioned framework uses a fixed margin to separate similar and dissimilar items, treating all non-groundtruth items as equally irrelevant. In this paper we propose to use a variable margin: we argue that varying the margin used during training based on how much relevant an item is to a given query, i.e. a relevance-based margin, easily improves the quality of the ranking lists measured through nDCG and mAP. We demonstrate the advantages of our technique using different models on EPIC-Kitchens-100 and YouCook2. We show that even if we carefully tuned the fixed margin, our technique (which does not have the margin as a hyper-parameter) would still achieve better performance. Finally, extensive ablation studies and qualitative analysis support the robustness of our approach. Code will be released at \url{https://github.com/aranciokov/RelevanceMargin-ICMR22}.

Galadrielle Humblot-Renaux, Sergio Escalera, Thomas B. Moeslund

While there has been a growing research interest in developing out-of-distribution (OOD) detection methods, there has been comparably little discussion around how these methods should be evaluated. Given their relevance for safe(r) AI, it is important to examine whether the basis for comparing OOD detection methods is consistent with practical needs. In this work, we take a closer look at the go-to metrics for evaluating OOD detection, and question the approach of exclusively reducing OOD detection to a binary classification task with little consideration for the detection threshold. We illustrate the limitations of current metrics (AUROC & its friends) and propose a new metric - Area Under the Threshold Curve (AUTC), which explicitly penalizes poor separation between ID and OOD samples. Scripts and data are available at https://github.com/glhr/beyond-auroc

Dustin Carrión-Ojeda, Hong Chen, Adrian El Baz et al.

We present the design and baseline results for a new challenge in the ChaLearn meta-learning series, accepted at NeurIPS'22, focusing on "cross-domain" meta-learning. Meta-learning aims to leverage experience gained from previous tasks to solve new tasks efficiently (i.e., with better performance, little training data, and/or modest computational resources). While previous challenges in the series focused on within-domain few-shot learning problems, with the aim of learning efficiently N-way k-shot tasks (i.e., N class classification problems with k training examples), this competition challenges the participants to solve "any-way" and "any-shot" problems drawn from various domains (healthcare, ecology, biology, manufacturing, and others), chosen for their humanitarian and societal impact. To that end, we created Meta-Album, a meta-dataset of 40 image classification datasets from 10 domains, from which we carve out tasks with any number of "ways" (within the range 2-20) and any number of "shots" (within the range 1-20). The competition is with code submission, fully blind-tested on the CodaLab challenge platform. The code of the winners will be open-sourced, enabling the deployment of automated machine learning solutions for few-shot image classification across several domains.

Mathilde Papillon, Mustafa Hajij, Helen Jenne et al.

This paper presents the computational challenge on topological deep learning that was hosted within the ICML 2023 Workshop on Topology and Geometry in Machine Learning. The competition asked participants to provide open-source implementations of topological neural networks from the literature by contributing to the python packages TopoNetX (data processing) and TopoModelX (deep learning). The challenge attracted twenty-eight qualifying submissions in its two-month duration. This paper describes the design of the challenge and summarizes its main findings.

Matthias Eisenmann, Annika Reinke, Vivienn Weru et al. · utoronto

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

Anthony Cioppa, Silvio Giancola, Vladimir Somers et al. · pku

The SoccerNet 2023 challenges were the third annual video understanding challenges organized by the SoccerNet team. For this third edition, the challenges were composed of seven vision-based tasks split into three main themes. The first theme, broadcast video understanding, is composed of three high-level tasks related to describing events occurring in the video broadcasts: (1) action spotting, focusing on retrieving all timestamps related to global actions in soccer, (2) ball action spotting, focusing on retrieving all timestamps related to the soccer ball change of state, and (3) dense video captioning, focusing on describing the broadcast with natural language and anchored timestamps. The second theme, field understanding, relates to the single task of (4) camera calibration, focusing on retrieving the intrinsic and extrinsic camera parameters from images. The third and last theme, player understanding, is composed of three low-level tasks related to extracting information about the players: (5) re-identification, focusing on retrieving the same players across multiple views, (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams, and (7) jersey number recognition, focusing on recognizing the jersey number of players from tracklets. Compared to the previous editions of the SoccerNet challenges, tasks (2-3-7) are novel, including new annotations and data, task (4) was enhanced with more data and annotations, and task (6) now focuses on end-to-end approaches. More information on the tasks, challenges, and leaderboards are available on https://www.soccer-net.org. Baselines and development kits can be found on https://github.com/SoccerNet.

Hao Fang, Ajian Liu, Jun Wan et al.

Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). In order to promote relevant research and fill this gap in the community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with 232 3D attacks (high-fidelity masks), 200 2D attacks (posters, portraits, and screens), and 2 adversarial attacks. In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: (1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. (2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. (3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.

Artur Xarles, Sergio Escalera, Thomas B. Moeslund et al.

Precise Event Spotting aims to localize fast-paced actions or events in videos with high temporal precision, a key task for applications in sports analytics, robotics, and autonomous systems. Existing methods typically process all frames uniformly, overlooking the inherent spatio-temporal redundancy in video data. This leads to redundant computation on non-informative regions while limiting overall efficiency. To remain tractable, they often spatially downsample inputs, losing fine-grained details crucial for precise localization. To address these limitations, we propose \textbf{AdaSpot}, a simple yet effective framework that processes low-resolution videos to extract global task-relevant features while adaptively selecting the most informative region-of-interest in each frame for high-resolution processing. The selection is performed via an unsupervised, task-aware strategy that maintains spatio-temporal consistency across frames and avoids the training instability of learnable alternatives. This design preserves essential fine-grained visual cues with a marginal computational overhead compared to low-resolution-only baselines, while remaining far more efficient than uniform high-resolution processing. Experiments on standard PES benchmarks demonstrate that \textbf{AdaSpot} achieves state-of-the-art performance under strict evaluation metrics (\eg, $+3.96$ and $+2.26$ mAP$@0$ frames on Tennis and FineDiving), while also maintaining strong results under looser metrics. Code is available at: \href{https://github.com/arturxe2/AdaSpot}{https://github.com/arturxe2/AdaSpot}.

German Barquero, Sergio Escalera, Cristina Palmero

Stochastic human motion prediction (HMP) has generally been tackled with generative adversarial networks and variational autoencoders. Most prior works aim at predicting highly diverse movements in terms of the skeleton joints' dispersion. This has led to methods predicting fast and motion-divergent movements, which are often unrealistic and incoherent with past motion. Such methods also neglect contexts that need to anticipate diverse low-range behaviors, or actions, with subtle joint displacements. To address these issues, we present BeLFusion, a model that, for the first time, leverages latent diffusion models in HMP to sample from a latent space where behavior is disentangled from pose and motion. As a result, diversity is encouraged from a behavioral perspective. Thanks to our behavior coupler's ability to transfer sampled behavior to ongoing motion, BeLFusion's predictions display a variety of behaviors that are significantly more realistic than the state of the art. To support it, we introduce two metrics, the Area of the Cumulative Motion Distribution, and the Average Pairwise Distance Error, which are correlated to our definition of realism according to a qualitative study with 126 participants. Finally, we prove BeLFusion's generalization power in a new cross-dataset scenario for stochastic HMP.

Silvio Giancola, Anthony Cioppa, Adrien Deliège et al.

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet.

Hugo Bertiche, Meysam Madadi, Sergio Escalera

We present a general framework for the garment animation problem through unsupervised deep learning inspired in physically based simulation. Existing trends in the literature already explore this possibility. Nonetheless, these approaches do not handle cloth dynamics. Here, we propose the first methodology able to learn realistic cloth dynamics unsupervisedly, and henceforth, a general formulation for neural cloth simulation. The key to achieve this is to adapt an existing optimization scheme for motion from simulation based methodologies to deep learning. Then, analyzing the nature of the problem, we devise an architecture able to automatically disentangle static and dynamic cloth subspaces by design. We will show how this improves model performance. Additionally, this opens the possibility of a novel motion augmentation technique that greatly improves generalization. Finally, we show it also allows to control the level of motion in the predictions. This is a useful, never seen before, tool for artists. We provide of detailed analysis of the problem to establish the bases of neural cloth simulation and guide future research into the specifics of this domain.

Joakim Bruslund Haurum, Sergio Escalera, Graham W. Taylor et al.

Since the introduction of the Vision Transformer (ViT), researchers have sought to make ViTs more efficient by removing redundant information in the processed tokens. While different methods have been explored to achieve this goal, we still lack understanding of the resulting reduction patterns and how those patterns differ across token reduction methods and datasets. To close this gap, we set out to understand the reduction patterns of 10 different token reduction methods using four image classification datasets. By systematically comparing these methods on the different classification tasks, we find that the Top-K pruning method is a surprisingly strong baseline. Through in-depth analysis of the different methods, we determine that: the reduction patterns are generally not consistent when varying the capacity of the backbone model, the reduction patterns of pruning-based methods significantly differ from fixed radial patterns, and the reduction patterns of pruning-based methods are correlated across classification datasets. Finally we report that the similarity of reduction patterns is a moderate-to-strong proxy for model performance. Project page at https://vap.aau.dk/tokens.

Neelu Madan, Àlex Pujol, Andreas Møgelmose et al.

Slot attention has emerged as a powerful framework for unsupervised object-centric learning, decomposing visual scenes into a small set of compact vector representations called \emph{slots}, each capturing a distinct region or object. However, these slots are learned in Euclidean space, which provides no geometric inductive bias for the hierarchical relationships that naturally structure visual scenes. In this work, we propose a simple post-hoc pipeline to project Euclidean slot embeddings onto the Lorentz hyperboloid of hyperbolic space, without modifying the underlying training pipeline. We construct five-level visual hierarchies directly from slot attention masks and analyse whether hyperbolic geometry reveals latent hierarchical structure that remains invisible in Euclidean space. Integrating our pipeline with SPOT (images), VideoSAUR (video), and SlotContrast (video), We find that hyperbolic projection exposes a consistent scene-level to object-level organisation, where coarse slots occupy greater manifold depth than fine slots, which is absent in Euclidean space. We further identify a "curvature--task tradeoff": low curvature ($c{=}0.2$) matches or outperforms Euclidean on parent slot retrieval, while moderate curvature ($c{=}0.5$) achieves better inter-level separation. Together, these findings suggest that slot representations already encode latent hierarchy that hyperbolic geometry reveals, motivating end-to-end hyperbolic training as a natural next step. Code and models are available at \href{https://github.com/NeeluMadan/HHS}{github.com/NeeluMadan/HHS}.

Ihsan Ullah, Dustin Carrión-Ojeda, Sergio Escalera et al.

We introduce Meta-Album, an image classification meta-dataset designed to facilitate few-shot learning, transfer learning, meta-learning, among other tasks. It includes 40 open datasets, each having at least 20 classes with 40 examples per class, with verified licences. They stem from diverse domains, such as ecology (fauna and flora), manufacturing (textures, vehicles), human actions, and optical character recognition, featuring various image scales (microscopic, human scales, remote sensing). All datasets are preprocessed, annotated, and formatted uniformly, and come in 3 versions (Micro $\subset$ Mini $\subset$ Extended) to match users' computational resources. We showcase the utility of the first 30 datasets on few-shot learning problems. The other 10 will be released shortly after. Meta-Album is already more diverse and larger (in number of datasets) than similar efforts, and we are committed to keep enlarging it via a series of competitions. As competitions terminate, their test data are released, thus creating a rolling benchmark, available through OpenML.org. Our website https://meta-album.github.io/ contains the source code of challenge winning methods, baseline methods, data loaders, and instructions for contributing either new datasets or algorithms to our expandable meta-dataset.

Swathikiran Sudhakaran, Sergio Escalera, Oswald Lanz

Convolutional Neural Networks are the de facto models for image recognition. However 3D CNNs, the straight forward extension of 2D CNNs for video recognition, have not achieved the same success on standard action recognition benchmarks. One of the main reasons for this reduced performance of 3D CNNs is the increased computational complexity requiring large scale annotated datasets to train them in scale. 3D kernel factorization approaches have been proposed to reduce the complexity of 3D CNNs. Existing kernel factorization approaches follow hand-designed and hard-wired techniques. In this paper we propose Gate-Shift-Fuse (GSF), a novel spatio-temporal feature extraction module which controls interactions in spatio-temporal decomposition and learns to adaptively route features through time and combine them in a data dependent manner. GSF leverages grouped spatial gating to decompose input tensor and channel weighting to fuse the decomposed tensors. GSF can be inserted into existing 2D CNNs to convert them into an efficient and high performing spatio-temporal feature extractor, with negligible parameter and compute overhead. We perform an extensive analysis of GSF using two popular 2D CNN families and achieve state-of-the-art or competitive performance on five standard action recognition benchmarks.

German Barquero, Johnny Núñez, Sergio Escalera et al.

Non-verbal social human behavior forecasting has increasingly attracted the interest of the research community in recent years. Its direct applications to human-robot interaction and socially-aware human motion generation make it a very attractive field. In this survey, we define the behavior forecasting problem for multiple interactive agents in a generic way that aims at unifying the fields of social signals prediction and human motion forecasting, traditionally separated. We hold that both problem formulations refer to the same conceptual problem, and identify many shared fundamental challenges: future stochasticity, context awareness, history exploitation, etc. We also propose a taxonomy that comprises methods published in the last 5 years in a very informative way and describes the current main concerns of the community with regard to this problem. In order to promote further research on this field, we also provide a summarised and friendly overview of audiovisual datasets featuring non-acted social interactions. Finally, we describe the most common metrics used in this task and their particular issues.

Hugo Bertiche, Niloy J. Mitra, Kuldeep Kulkarni et al.

Cinemagraphs are short looping videos created by adding subtle motions to a static image. This kind of media is popular and engaging. However, automatic generation of cinemagraphs is an underexplored area and current solutions require tedious low-level manual authoring by artists. In this paper, we present an automatic method that allows generating human cinemagraphs from single RGB images. We investigate the problem in the context of dressed humans under the wind. At the core of our method is a novel cyclic neural network that produces looping cinemagraphs for the target loop duration. To circumvent the problem of collecting real data, we demonstrate that it is possible, by working in the image normal space, to learn garment motion dynamics on synthetic data and generalize to real data. We evaluate our method on both synthetic and real data and demonstrate that it is possible to create compelling and plausible cinemagraphs from single RGB images.

Dong Wang, Jia Guo, Qiqi Shao et al.

Face anti-spoofing (FAS) is an essential mechanism for safeguarding the integrity of automated face recognition systems. Despite substantial advancements, the generalization of existing approaches to real-world applications remains challenging. This limitation can be attributed to the scarcity and lack of diversity in publicly available FAS datasets, which often leads to overfitting during training or saturation during testing. In terms of quantity, the number of spoof subjects is a critical determinant. Most datasets comprise fewer than 2,000 subjects. With regard to diversity, the majority of datasets consist of spoof samples collected in controlled environments using repetitive, mechanical processes. This data collection methodology results in homogenized samples and a dearth of scenario diversity. To address these shortcomings, we introduce the Wild Face Anti-Spoofing (WFAS) dataset, a large-scale, diverse FAS dataset collected in unconstrained settings. Our dataset encompasses 853,729 images of 321,751 spoof subjects and 529,571 images of 148,169 live subjects, representing a substantial increase in quantity. Moreover, our dataset incorporates spoof data obtained from the internet, spanning a wide array of scenarios and various commercial sensors, including 17 presentation attacks (PAs) that encompass both 2D and 3D forms. This novel data collection strategy markedly enhances FAS data diversity. Leveraging the WFAS dataset and Protocol 1 (Known-Type), we host the Wild Face Anti-Spoofing Challenge at the CVPR2023 workshop. Additionally, we meticulously evaluate representative methods using Protocol 1 and Protocol 2 (Unknown-Type). Through an in-depth examination of the challenge outcomes and benchmark baselines, we provide insightful analyses and propose potential avenues for future research. The dataset is released under Insightface.

Hao Fang, Ajian Liu, Jun Wan et al.

Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, most of the studies lacked consideration of long-distance scenarios. Specifically, compared with FAS in traditional scenes such as phone unlocking, face payment, and self-service security inspection, FAS in long-distance such as station squares, parks, and self-service supermarkets are equally important, but it has not been sufficiently explored yet. In order to fill this gap in the FAS community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask). SuHiFiMask contains $10,195$ videos from $101$ subjects of different age groups, which are collected by $7$ mainstream surveillance cameras. Based on this dataset and protocol-$3$ for evaluating the robustness of the algorithm under quality changes, we organized a face presentation attack detection challenge in surveillance scenarios. It attracted 180 teams for the development phase with a total of 37 teams qualifying for the final round. The organization team re-verified and re-ran the submitted code and used the results as the final ranking. In this paper, we present an overview of the challenge, including an introduction to the dataset used, the definition of the protocol, the evaluation metrics, and the announcement of the competition results. Finally, we present the top-ranked algorithms and the research ideas provided by the competition for attack detection in long-range surveillance scenarios.

Razieh Rastgoo, Kourosh Kiani, Sergio Escalera

. Continuous Sign Language Recognition (CSLR) is a long challenging task in Computer Vision due to the difficulties in detecting the explicit boundaries between the words in a sign sentence. To deal with this challenge, we propose a two-stage model. In the first stage, the predictor model, which includes a combination of CNN, SVD, and LSTM, is trained with the isolated signs. In the second stage, we apply a post-processing algorithm to the Softmax outputs obtained from the first part of the model in order to separate the isolated signs in the continuous signs. While the proposed model is trained on the isolated sign classes with similar frame numbers, it is evaluated on the continuous sign videos with a different frame length per each isolated sign class. Due to the lack of a large dataset, including both the sign sequences and the corresponding isolated signs, two public datasets in Isolated Sign Language Recognition (ISLR), RKS-PERSIANSIGN and ASLLVD, are used for evaluation. Results of the continuous sign videos confirm the efficiency of the proposed model to deal with isolated sign boundaries detection.

German Barquero, Johnny Núñez, Zhen Xu et al.

Human behavior forecasting during human-human interactions is of utmost importance to provide robotic or virtual agents with social intelligence. This problem is especially challenging for scenarios that are highly driven by interpersonal dynamics. In this work, we present the first systematic comparison of state-of-the-art approaches for behavior forecasting. To do so, we leverage whole-body annotations (face, body, and hands) from the very recently released UDIVA v0.5, which features face-to-face dyadic interactions. Our best attention-based approaches achieve state-of-the-art performance in UDIVA v0.5. We show that by autoregressively predicting the future with methods trained for the short-term future (<400ms), we outperform the baselines even for a considerably longer-term future (up to 2s). We also show that this finding holds when highly noisy annotations are used, which opens new horizons towards the use of weakly-supervised learning. Combined with large-scale datasets, this may help boost the advances in this field.

Arya Farkhondeh, Cristina Palmero, Simone Scardapane et al.

Recent joint embedding-based self-supervised methods have surpassed standard supervised approaches on various image recognition tasks such as image classification. These self-supervised methods aim at maximizing agreement between features extracted from two differently transformed views of the same image, which results in learning an invariant representation with respect to appearance and geometric image transformations. However, the effectiveness of these approaches remains unclear in the context of gaze estimation, a structured regression task that requires equivariance under geometric transformations (e.g., rotations, horizontal flip). In this work, we propose SwAT, an equivariant version of the online clustering-based self-supervised approach SwAV, to learn more informative representations for gaze estimation. We demonstrate that SwAT, with ResNet-50 and supported with uncurated unlabeled face images, outperforms state-of-the-art gaze estimation methods and supervised baselines in various experiments. In particular, we achieve up to 57% and 25% improvements in cross-dataset and within-dataset evaluation tasks on existing benchmarks (ETH-XGaze, Gaze360, and MPIIFaceGaze).

Rubén Ballester, Xavier Arnal Clemente, Carles Casacuberta et al.

Understanding how neural networks generalize on unseen data is crucial for designing more robust and reliable models. In this paper, we study the generalization gap of neural networks using methods from topological data analysis. For this purpose, we compute homological persistence diagrams of weighted graphs constructed from neuron activation correlations after a training phase, aiming to capture patterns that are linked to the generalization capacity of the network. We compare the usefulness of different numerical summaries from persistence diagrams and show that a combination of some of them can accurately predict and partially explain the generalization gap without the need of a test set. Evaluation on two computer vision recognition tasks (CIFAR10 and SVHN) shows competitive generalization gap prediction when compared against state-of-the-art methods.

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.

Joakim Bruslund Haurum, Sergio Escalera, Graham W. Taylor et al.

We present Agglomerative Token Clustering (ATC), a novel token merging method that consistently outperforms previous token merging and pruning methods across image classification, image synthesis, and object detection & segmentation tasks. ATC merges clusters through bottom-up hierarchical clustering, without the introduction of extra learnable parameters. We find that ATC achieves state-of-the-art performance across all tasks, and can even perform on par with prior state-of-the-art when applied off-the-shelf, i.e. without fine-tuning. ATC is particularly effective when applied with low keep rates, where only a small fraction of tokens are kept and retaining task performance is especially difficult.

Rubén Ballester, Carles Casacuberta, Sergio Escalera

We propose a novel way to improve the generalisation capacity of deep learning models by reducing high correlations between neurons. For this, we present two regularisation terms computed from the weights of a minimum spanning tree of the clique whose vertices are the neurons of a given network (or a sample of those), where weights on edges are correlation dissimilarities. We provide an extensive set of experiments to validate the effectiveness of our terms, showing that they outperform popular ones. Also, we demonstrate that naive minimisation of all correlations between neurons obtains lower accuracies than our regularisation terms, suggesting that redundancies play a significant role in artificial neural networks, as evidenced by some studies in neuroscience for real networks. We include a proof of differentiability of our regularisers, thus developing the first effective topological persistence-based regularisation terms that consider the whole set of neurons and that can be applied to a feedforward architecture in any deep learning task such as classification, data generation, or regression.

Razieh Rastgoo, Kourosh Kiani, Sergio Escalera

Continuous Hand Gesture Recognition (CHGR) has been extensively studied by researchers in the last few decades. Recently, one model has been presented to deal with the challenge of the boundary detection of isolated gestures in a continuous gesture video [17]. To enhance the model performance and also replace the handcrafted feature extractor in the presented model in [17], we propose a GCN model and combine it with the stacked Bi-LSTM and Attention modules to push the temporal information in the video stream. Considering the breakthroughs of GCN models for skeleton modality, we propose a two-layer GCN model to empower the 3D hand skeleton features. Finally, the class probabilities of each isolated gesture are fed to the post-processing module, borrowed from [17]. Furthermore, we replace the anatomical graph structure with some non-anatomical graph structures. Due to the lack of a large dataset, including both the continuous gesture sequences and the corresponding isolated gestures, three public datasets in Dynamic Hand Gesture Recognition (DHGR), RKS-PERSIANSIGN, and ASLVID, are used for evaluation. Experimental results show the superiority of the proposed model in dealing with isolated gesture boundaries detection in continuous gesture sequences

Alex Falcon, Giuseppe Serra, Sergio Escalera et al.

This report presents the technical details of our submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2022. To participate in the challenge, we designed an ensemble consisting of different models trained with two recently developed relevance-augmented versions of the widely used triplet loss. Our submission, visible on the public leaderboard, obtains an average score of 61.02% nDCG and 49.77% mAP.

Hunor Laczkó, Meysam Madadi, Sergio Escalera et al.

RGB cloth generation has been deeply studied in the related literature, however, 3D garment generation remains an open problem. In this paper, we build a conditional variational autoencoder for 3D garment generation and draping. We propose a pyramid network to add garment details progressively in a canonical space, i.e. unposing and unshaping the garments w.r.t. the body. We study conditioning the network on surface normal UV maps, as an intermediate representation, which is an easier problem to optimize than 3D coordinates. Our results on two public datasets, CLOTH3D and CAPE, show that our model is robust, controllable in terms of detail generation by the use of multi-resolution pyramids, and achieves state-of-the-art results that can highly generalize to unseen garments, poses, and shapes even when training with small amounts of data.

Cristina Palmero, Mikel deVelasco, Mohamed Amine Hmani et al.

The EMPATHIC project aimed to design an emotionally expressive virtual coach capable of engaging healthy seniors to improve well-being and promote independent aging. One of the core aspects of the system is its human sensing capabilities, allowing for the perception of emotional states to provide a personalized experience. This paper outlines the development of the emotion expression recognition module of the virtual coach, encompassing data collection, annotation design, and a first methodological approach, all tailored to the project requirements. With the latter, we investigate the role of various modalities, individually and combined, for discrete emotion expression recognition in this context: speech from audio, and facial expressions, gaze, and head dynamics from video. The collected corpus includes users from Spain, France, and Norway, and was annotated separately for the audio and video channels with distinct emotional labels, allowing for a performance comparison across cultures and label types. Results confirm the informative power of the modalities studied for the emotional categories considered, with multimodal methods generally outperforming others (around 68% accuracy with audio labels and 72-74% with video labels). The findings are expected to contribute to the limited literature on emotion recognition applied to older adults in conversational human-machine interaction.

Loc-Phat Truong, Meysam Madadi, Sergio Escalera

Virtual Try-On (VTON) has seen rapid advancements, providing a strong foundation for generative fashion tasks. However, the inverse problem, Virtual Try-Off (VTOFF)-aimed at reconstructing the canonical garment from a draped-on image-remains a less understood domain, distinct from the heavily researched field of VTON. In this work, we seek to establish a robust architectural foundation for VTOFF by studying and adapting various diffusion-based strategies from VTON and general Latent Diffusion Models (LDMs). We focus our investigation on the Dual-UNet Diffusion Model architecture and analyze three axes of design: (i) Generation Backbone: comparing Stable Diffusion variants; (ii) Conditioning: ablating different mask designs, masked/unmasked inputs for image conditioning, and the utility of high-level semantic features; and (iii) Losses and Training Strategies: evaluating the impact of the auxiliary attention-based loss, perceptual objectives and multi-stage curriculum schedules. Extensive experiments reveal trade-offs across various configuration options. Evaluated on VITON-HD and DressCode datasets, our framework achieves state-of-the-art performance with a drop of 9.5\% on the primary metric DISTS and competitive performance on LPIPS, FID, KID, and SSIM, providing both stronger baselines and insights to guide future Virtual Try-Off research.

Senyuan Shi, Hao Tan, Zichang Tan et al.

The rapid evolution of generative models has precipitated a proliferation of fabricated content, posing significant challenges to existing Synthetic Image Detection (SID) methods. Capitalizing on advancements in vision-language models (e.g., CLIP), recent attempts have leveraged learnable textual prompts to identify synthetic images. However, they still leverage static prompt as a fixed boundary for real and fake images, failing to adapt to the varying types of forgery that emerge during inference. To overcome this issue, we propose **HydraPrompt**, an asymmetric prompting framework that dynamically adjusts the category centers by aligning with fine-grained image cues. Specifically, we propose an Asymmetric Prompt Adapter (**APA**): (1) for authentic category, we introduce a single set of prompts to capture the consistent representative patterns, which serves as a unified anchor for real content. While (2) for fake category, we construct sample-adaptive prompts that specialize in capturing diverse cues from different samples, enabling adaptive modeling of forgery image variations. To increase pronounced discriminability within different synthetic images, we further introduce a Conditional Supervised Contrastive (**CSC**) objective, which compacts the authentic representations while capturing fine-grained forgery clues. Extensive experiments on popular SID benchmarks demonstrate the state-of-the-art performance of our framework.

David Pujol-Perich, Albert Clapés, Dima Damen et al.

In this work, we investigate the degradation of existing VMR methods, particularly of DETR architectures, when trained on caption-based queries but evaluated on search queries. For this, we introduce three benchmarks by modifying the textual queries in three public VMR datasets -- i.e., HD-EPIC, YouCook2 and ActivityNet-Captions. Our analysis reveals two key generalization challenges: (i) A language gap, arising from the linguistic under-specification of search queries, and (ii) a multi-moment gap, caused by the shift from single-moment to multi-moment queries. We also identify a critical issue in these architectures -- an active decoder-query collapse -- as a primary cause of the poor generalization to multi-moment instances. We mitigate this issue with architectural modifications that effectively increase the number of active decoder queries. Extensive experiments demonstrate that our approach improves performance on search queries by up to 14.82% mAP_m, and up to 21.83% mAP_m on multi-moment search queries. The code, models and data are available in the project webpage: https://davidpujol.github.io/beyond-vmr/

Artur Xarles, Sergio Escalera, Thomas B. Moeslund et al.

In this paper, we introduce T-DEED, a Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in sports videos. T-DEED addresses multiple challenges in the task, including the need for discriminability among frame representations, high output temporal resolution to maintain prediction precision, and the necessity to capture information at different temporal scales to handle events with varying dynamics. It tackles these challenges through its specifically designed architecture, featuring an encoder-decoder for leveraging multiple temporal scales and achieving high output temporal resolution, along with temporal modules designed to increase token discriminability. Leveraging these characteristics, T-DEED achieves SOTA performance on the FigureSkating and FineDiving datasets. Code is available at https://github.com/arturxe2/T-DEED.

Siyang Song, Micol Spitale, Cheng Luo et al.

In dyadic interactions, humans communicate their intentions and state of mind using verbal and non-verbal cues, where multiple different facial reactions might be appropriate in response to a specific speaker behaviour. Then, how to develop a machine learning (ML) model that can automatically generate multiple appropriate, diverse, realistic and synchronised human facial reactions from an previously unseen speaker behaviour is a challenging task. Following the successful organisation of the first REACT challenge (REACT 2023), this edition of the challenge (REACT 2024) employs a subset used by the previous challenge, which contains segmented 30-secs dyadic interaction clips originally recorded as part of the NOXI and RECOLA datasets, encouraging participants to develop and benchmark Machine Learning (ML) models that can generate multiple appropriate facial reactions (including facial image sequences and their attributes) given an input conversational partner's stimulus under various dyadic video conference scenarios. This paper presents: (i) the guidelines of the REACT 2024 challenge; (ii) the dataset utilized in the challenge; and (iii) the performance of the baseline systems on the two proposed sub-challenges: Offline Multiple Appropriate Facial Reaction Generation and Online Multiple Appropriate Facial Reaction Generation, respectively. The challenge baseline code is publicly available at https://github.com/reactmultimodalchallenge/baseline_react2024.

Amir Ansari, Jana Subirana, Bruna Silva et al.

Personality is a complex, hierarchical construct typically assessed through item-level questionnaires aggregated into broad trait scores. Personality recognition models aim to infer personality traits from different sources of behavioral data. However, reliance on broad trait scores as ground truth, combined with limited training data, poses challenges for generalization, as similar trait scores can manifest through diverse, context dependent behaviors. In this work, we explore the predictive impact of the more granular hierarchical levels of the Big-Five Personality Model, facets and nuances, to enhance personality recognition from audiovisual interaction data. Using the UDIVA v0.5 dataset, we trained a transformer-based model including cross-modal (audiovisual) and cross-subject (dyad-aware) attention mechanisms. Results show that nuance-level models consistently outperform facet and trait-level models, reducing mean squared error by up to 74% across interaction scenarios.

Galadrielle Humblot-Renaux, Sergio Escalera, Thomas B. Moeslund

The ability to detect unfamiliar or unexpected images is essential for safe deployment of computer vision systems. In the context of classification, the task of detecting images outside of a model's training domain is known as out-of-distribution (OOD) detection. While there has been a growing research interest in developing post-hoc OOD detection methods, there has been comparably little discussion around how these methods perform when the underlying classifier is not trained on a clean, carefully curated dataset. In this work, we take a closer look at 20 state-of-the-art OOD detection methods in the (more realistic) scenario where the labels used to train the underlying classifier are unreliable (e.g. crowd-sourced or web-scraped labels). Extensive experiments across different datasets, noise types & levels, architectures and checkpointing strategies provide insights into the effect of class label noise on OOD detection, and show that poor separation between incorrectly classified ID samples vs. OOD samples is an overlooked yet important limitation of existing methods. Code: https://github.com/glhr/ood-labelnoise

Pablo Ruiz-Ponce, Sergio Escalera, José García-Rodríguez et al.

Generating realistic human-human interactions is a challenging task that requires not only high-quality individual body and hand motions, but also coherent coordination among all interactants. Due to limitations in available data and increased learning complexity, previous methods tend to ignore hand motions, limiting the realism and expressivity of the interactions. Additionally, current diffusion-based approaches generate entire motion sequences simultaneously, limiting their ability to capture the reactive and adaptive nature of human interactions. To address these limitations, we introduce Interact2Ar, the first end-to-end text-conditioned autoregressive diffusion model for generating full-body, human-human interactions. Interact2Ar incorporates detailed hand kinematics through dedicated parallel branches, enabling high-fidelity full-body generation. Furthermore, we introduce an autoregressive pipeline coupled with a novel memory technique that facilitates adaptation to the inherent variability of human interactions using efficient large context windows. The adaptability of our model enables a series of downstream applications, including temporal motion composition, real-time adaptation to disturbances, and extension beyond dyadic to multi-person scenarios. To validate the generated motions, we introduce a set of robust evaluators and extended metrics designed specifically for assessing full-body interactions. Through quantitative and qualitative experiments, we demonstrate the state-of-the-art performance of Interact2Ar.

Siyang Song, Micol Spitale, Xiangyu Kong et al.

In dyadic interactions, a broad spectrum of human facial reactions might be appropriate for responding to each human speaker behaviour. Following the successful organisation of the REACT 2023 and REACT 2024 challenges, we are proposing the REACT 2025 challenge encouraging the development and benchmarking of Machine Learning (ML) models that can be used to generate multiple appropriate, diverse, realistic and synchronised human-style facial reactions expressed by human listeners in response to an input stimulus (i.e., audio-visual behaviours expressed by their corresponding speakers). As a key of the challenge, we provide challenge participants with the first natural and large-scale multi-modal MAFRG dataset (called MARS) recording 137 human-human dyadic interactions containing a total of 2856 interaction sessions covering five different topics. In addition, this paper also presents the challenge guidelines and the performance of our baselines on the two proposed sub-challenges: Offline MAFRG and Online MAFRG, respectively. The challenge baseline code is publicly available at https://github.com/reactmultimodalchallenge/baseline_react2025

Zhi Rao, Yucheng Zhou, Benjia Zhou et al.

Gloss-free sign language translation (SLT) is hindered by two key challenges: **inadequate sign representation** that fails to capture nuanced visual cues, and **sentence-level semantic misalignment** in current LLM-based methods, which limits translation quality. To address these issues, we propose a three-stage **r**einforcing **v**ision-**l**anguage **f**ramework (**RVLF**). We build a large vision-language model (LVLM) specifically designed for sign language, and then combine it with reinforcement learning (RL) to adaptively enhance translation performance. First, for a sufficient representation of sign language, RVLF introduces an effective semantic representation learning mechanism that fuses skeleton-based motion cues with semantically rich visual features extracted via DINOv2, followed by instruction tuning to obtain a strong SLT-SFT baseline. Then, to improve sentence-level semantic misalignment, we introduce a GRPO-based optimization strategy that fine-tunes the SLT-SFT model with a reward function combining translation fidelity (BLEU) and sentence completeness (ROUGE), yielding the optimized model termed SLT-GRPO. Our conceptually simple framework yields substantial gains under the gloss-free SLT setting without pre-training on any external large-scale sign language datasets, improving BLEU-4 scores by +5.1, +1.11, +1.4, and +1.61 on the CSL-Daily, PHOENIX-2014T, How2Sign, and OpenASL datasets, respectively. To the best of our knowledge, this is the first work to incorporate GRPO into SLT. Extensive experiments and ablation studies validate the effectiveness of GRPO-based optimization in enhancing both translation quality and semantic consistency.

Oriol Rabasseda, Zenjie Li, Kamal Nasrollahi et al.

Automatic identification of events and recurrent behavior analysis are critical for video surveillance. However, most existing content-based video retrieval benchmarks focus on scene-level similarity and do not evaluate the action discrimination required in surveillance. To address this gap, we introduce SOVABench (Surveillance Opposite Vehicle Actions Benchmark), a real-world retrieval benchmark built from surveillance footage and centered on vehicle-related actions. SOVABench defines two evaluation protocols (inter-pair and intra-pair) to assess cross-action discrimination and temporal direction understanding. Although action distinctions are generally intuitive for human observers, our experiments show that they remain challenging for state-of-the-art vision and multimodal models. Leveraging the visual reasoning and instruction-following capabilities of Multimodal Large Language Models (MLLMs), we present a training-free framework for producing interpretable embeddings from MLLM-generated descriptions for both images and videos. The framework achieves strong performance on SOVABench as well as on several spatial and counting benchmarks where contrastive Vision-Language Models often fail. The code, annotations, and instructions to construct the benchmark are publicly available.

David Pujol-Perich, Sergio Escalera, Albert Clapés

Video Temporal Grounding (VTG) involves Moment Retrieval (MR) and Highlight Detection (HD) based on textual queries. For this, most methods rely solely on final-layer features of frozen large pre-trained backbones, limiting their adaptability to new domains. While full fine-tuning is often impractical, parameter-efficient fine-tuning -- and particularly side-tuning (ST) -- has emerged as an effective alternative. However, prior ST approaches this problem from a frame-level refinement perspective, overlooking the inherent sparse nature of MR. To address this, we propose the Sparse-Dense Side-Tuner (SDST), the first anchor-free ST architecture for VTG. We also introduce the Reference-based Deformable Self-Attention, a novel mechanism that enhances the context modeling of the deformable attention -- a key limitation of existing anchor-free methods. Additionally, we present the first effective integration of InternVideo2 backbone into an ST framework, showing its profound implications in performance. Overall, our method significantly improves existing ST methods, achieving highly competitive or SOTA results on QVHighlights, TACoS, and Charades-STA, while reducing up to a 73% the parameter count w.r.t. the existing SOTA methods. The code is publicly accessible at https://github.com/davidpujol/SDST.

Àlex Pujol Vidal, Sergio Escalera, Kamal Nasrollahi et al.

Machine unlearning methods have become increasingly important for selective concept removal in large pre-trained models. While recent work has explored unlearning in Euclidean contrastive vision-language models, the effectiveness of concept removal in hyperbolic spaces remains unexplored. This paper investigates machine unlearning in hyperbolic contrastive learning by adapting Alignment Calibration to MERU, a model that embeds images and text in hyperbolic space to better capture semantic hierarchies. Through systematic experiments and ablation studies, we demonstrate that hyperbolic geometry offers distinct advantages for concept removal, achieving near perfect forgetting with reasonable performance on retained concepts, particularly when scaling to multiple concept removal. Our approach introduces hyperbolic-specific components including entailment calibration and norm regularization that leverage the unique properties of hyperbolic space. Comparative analysis with Euclidean models reveals fundamental differences in unlearning dynamics, with hyperbolic unlearning reorganizing the semantic hierarchy while Euclidean approaches merely disconnect cross-modal associations. These findings not only advance machine unlearning techniques but also provide insights into the geometric properties that influence concept representation and removal in multimodal models. Source code available at https://github.com/alex-pv01/HAC

Mohamad Dalal, Artur Xarles, Anthony Cioppa et al.

Artificial intelligence has revolutionized the way we analyze sports videos, whether to understand the actions of games in long untrimmed videos or to anticipate the player's motion in future frames. Despite these efforts, little attention has been given to anticipating game actions before they occur. In this work, we introduce the task of action anticipation for football broadcast videos, which consists in predicting future actions in unobserved future frames, within a five- or ten-second anticipation window. To benchmark this task, we release a new dataset, namely the SoccerNet Ball Action Anticipation dataset, based on SoccerNet Ball Action Spotting. Additionally, we propose a Football Action ANticipation TRAnsformer (FAANTRA), a baseline method that adapts FUTR, a state-of-the-art action anticipation model, to predict ball-related actions. To evaluate action anticipation, we introduce new metrics, including mAP@$δ$, which evaluates the temporal precision of predicted future actions, as well as mAP@$\infty$, which evaluates their occurrence within the anticipation window. We also conduct extensive ablation studies to examine the impact of various task settings, input configurations, and model architectures. Experimental results highlight both the feasibility and challenges of action anticipation in football videos, providing valuable insights into the design of predictive models for sports analytics. By forecasting actions before they unfold, our work will enable applications in automated broadcasting, tactical analysis, and player decision-making. Our dataset and code are publicly available at https://github.com/MohamadDalal/FAANTRA.

German Barquero, Nadine Bertsch, Manojkumar Marramreddy et al.

In extended reality (XR), generating full-body motion of the users is important to understand their actions, drive their virtual avatars for social interaction, and convey a realistic sense of presence. While prior works focused on spatially sparse and always-on input signals from motion controllers, many XR applications opt for vision-based hand tracking for reduced user friction and better immersion. Compared to controllers, hand tracking signals are less accurate and can even be missing for an extended period of time. To handle such unreliable inputs, we present Rolling Prediction Model (RPM), an online and real-time approach that generates smooth full-body motion from temporally and spatially sparse input signals. Our model generates 1) accurate motion that matches the inputs (i.e., tracking mode) and 2) plausible motion when inputs are missing (i.e., synthesis mode). More importantly, RPM generates seamless transitions from tracking to synthesis, and vice versa. To demonstrate the practical importance of handling noisy and missing inputs, we present GORP, the first dataset of realistic sparse inputs from a commercial virtual reality (VR) headset with paired high quality body motion ground truth. GORP provides >14 hours of VR gameplay data from 28 people using motion controllers (spatially sparse) and hand tracking (spatially and temporally sparse). We benchmark RPM against the state of the art on both synthetic data and GORP to highlight how we can bridge the gap for real-world applications with a realistic dataset and by handling unreliable input signals. Our code, pretrained models, and GORP dataset are available in the project webpage.

Galadrielle Humblot-Renaux, Gianni Franchi, Sergio Escalera et al.

Out-of-distribution (OOD) detection is an important building block in trustworthy image recognition systems as unknown classes may arise at test-time. OOD detection methods typically revolve around a single classifier, leading to a split in the research field between the classical supervised setting (e.g. ResNet18 classifier trained on CIFAR100) vs. the zero-shot setting (class names fed as prompts to CLIP). In both cases, an overarching challenge is that the OOD detection performance is implicitly constrained by the classifier's capabilities on in-distribution (ID) data. In this work, we show that given a little open-mindedness from both ends, remarkable OOD detection can be achieved by instead creating a heterogeneous ensemble - COOkeD combines the predictions of a closed-world classifier trained end-to-end on a specific dataset, a zero-shot CLIP classifier, and a linear probe classifier trained on CLIP image features. While bulky at first sight, this approach is modular, post-hoc and leverages the availability of pre-trained VLMs, thus introduces little overhead compared to training a single standard classifier. We evaluate COOkeD on popular CIFAR100 and ImageNet benchmarks, but also consider more challenging, realistic settings ranging from training-time label noise, to test-time covariate shift, to zero-shot shift which has been previously overlooked. Despite its simplicity, COOkeD achieves state-of-the-art performance and greater robustness compared to both classical and CLIP-based OOD detection methods. Code is available at https://github.com/glhr/COOkeD

Zhengying Liu, Adrien Pavao, Zhen Xu et al.

This paper reports the results and post-challenge analyses of ChaLearn's AutoDL challenge series, which helped sorting out a profusion of AutoML solutions for Deep Learning (DL) that had been introduced in a variety of settings, but lacked fair comparisons. All input data modalities (time series, images, videos, text, tabular) were formatted as tensors and all tasks were multi-label classification problems. Code submissions were executed on hidden tasks, with limited time and computational resources, pushing solutions that get results quickly. In this setting, DL methods dominated, though popular Neural Architecture Search (NAS) was impractical. Solutions relied on fine-tuned pre-trained networks, with architectures matching data modality. Post-challenge tests did not reveal improvements beyond the imposed time limit. While no component is particularly original or novel, a high level modular organization emerged featuring a "meta-learner", "data ingestor", "model selector", "model/learner", and "evaluator". This modularity enabled ablation studies, which revealed the importance of (off-platform) meta-learning, ensembling, and efficient data management. Experiments on heterogeneous module combinations further confirm the (local) optimality of the winning solutions. Our challenge legacy includes an ever-lasting benchmark (http://autodl.chalearn.org), the open-sourced code of the winners, and a free "AutoDL self-service".

Zhen Xu, Sergio Escalera, Isabelle Guyon et al.

Obtaining standardized crowdsourced benchmark of computational methods is a major issue in data science communities. Dedicated frameworks enabling fair benchmarking in a unified environment are yet to be developed. Here we introduce Codabench, an open-source, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks. A public instance of Codabench (https://www.codabench.org/) is open to everyone, free of charge, and allows benchmark organizers to compare fairly submissions, under the same setting (software, hardware, data, algorithms), with custom protocols and data formats. Codabench has unique features facilitating the organization of benchmarks flexibly, easily and reproducibly, such as the possibility of re-using templates of benchmarks, and supplying compute resources on-demand. Codabench has been used internally and externally on various applications, receiving more than 130 users and 2500 submissions. As illustrative use cases, we introduce 4 diverse benchmarks covering Graph Machine Learning, Cancer Heterogeneity, Clinical Diagnosis and Reinforcement Learning.