Fabio Galasso

Cristiano Saltori, Fabio Galasso, Giuseppe Fiameni et al.

3D LiDAR semantic segmentation is fundamental for autonomous driving. Several Unsupervised Domain Adaptation (UDA) methods for point cloud data have been recently proposed to improve model generalization for different sensors and environments. Researchers working on UDA problems in the image domain have shown that sample mixing can mitigate domain shift. We propose a new approach of sample mixing for point cloud UDA, namely Compositional Semantic Mix (CoSMix), the first UDA approach for point cloud segmentation based on sample mixing. CoSMix consists of a two-branch symmetric network that can process labelled synthetic data (source) and real-world unlabelled point clouds (target) concurrently. Each branch operates on one domain by mixing selected pieces of data from the other one, and by using the semantic information derived from source labels and target pseudo-labels. We evaluate CoSMix on two large-scale datasets, showing that it outperforms state-of-the-art methods by a large margin. Our code is available at https://github.com/saltoricristiano/cosmix-uda.

Cristiano Saltori, Evgeny Krivosheev, Stéphane Lathuilière et al.

3D point cloud semantic segmentation is fundamental for autonomous driving. Most approaches in the literature neglect an important aspect, i.e., how to deal with domain shift when handling dynamic scenes. This can significantly hinder the navigation capabilities of self-driving vehicles. This paper advances the state of the art in this research field. Our first contribution consists in analysing a new unexplored scenario in point cloud segmentation, namely Source-Free Online Unsupervised Domain Adaptation (SF-OUDA). We experimentally show that state-of-the-art methods have a rather limited ability to adapt pre-trained deep network models to unseen domains in an online manner. Our second contribution is an approach that relies on adaptive self-training and geometric-feature propagation to adapt a pre-trained source model online without requiring either source data or target labels. Our third contribution is to study SF-OUDA in a challenging setup where source data is synthetic and target data is point clouds captured in the real world. We use the recent SynLiDAR dataset as a synthetic source and introduce two new synthetic (source) datasets, which can stimulate future synthetic-to-real autonomous driving research. Our experiments show the effectiveness of our segmentation approach on thousands of real-world point clouds. Code and synthetic datasets are available at https://github.com/saltoricristiano/gipso-sfouda.

Luca Franco, Paolo Mandica, Bharti Munjal et al.

Self-paced learning has been beneficial for tasks where some initial knowledge is available, such as weakly supervised learning and domain adaptation, to select and order the training sample sequence, from easy to complex. However its applicability remains unexplored in unsupervised learning, whereby the knowledge of the task matures during training. We propose a novel HYperbolic Self-Paced model (HYSP) for learning skeleton-based action representations. HYSP adopts self-supervision: it uses data augmentations to generate two views of the same sample, and it learns by matching one (named online) to the other (the target). We propose to use hyperbolic uncertainty to determine the algorithmic learning pace, under the assumption that less uncertain samples should be more strongly driving the training, with a larger weight and pace. Hyperbolic uncertainty is a by-product of the adopted hyperbolic neural networks, it matures during training and it comes with no extra cost, compared to the established Euclidean SSL framework counterparts. When tested on three established skeleton-based action recognition datasets, HYSP outperforms the state-of-the-art on PKU-MMD I, as well as on 2 out of 3 downstream tasks on NTU-60 and NTU-120. Additionally, HYSP only uses positive pairs and bypasses therefore the complex and computationally-demanding mining procedures required for the negatives in contrastive techniques. Code is available at https://github.com/paolomandica/HYSP.

Laura Laurenti, Elisa Tinti, Fabio Galasso et al.

Earthquake forecasting and prediction have long and in some cases sordid histories but recent work has rekindled interest based on advances in early warning, hazard assessment for induced seismicity and successful prediction of laboratory earthquakes. In the lab, frictional stick-slip events provide an analog for earthquakes and the seismic cycle. Labquakes are ideal targets for machine learning (ML) because they can be produced in long sequences under controlled conditions. Recent works show that ML can predict several aspects of labquakes using fault zone acoustic emissions. Here, we generalize these results and explore deep learning (DL) methods for labquake prediction and autoregressive (AR) forecasting. DL improves existing ML methods of labquake prediction. AR methods allow forecasting at future horizons via iterative predictions. We demonstrate that DL models based on Long-Short Term Memory (LSTM) and Convolution Neural Networks predict labquakes under several conditions, and that fault zone stress can be predicted with fidelity, confirming that acoustic energy is a fingerprint of fault zone stress. We predict also time to start of failure (TTsF) and time to the end of Failure (TTeF) for labquakes. Interestingly, TTeF is successfully predicted in all seismic cycles, while the TTsF prediction varies with the amount of preseismic fault creep. We report AR methods to forecast the evolution of fault stress using three sequence modeling frameworks: LSTM, Temporal Convolution Network and Transformer Network. AR forecasting is distinct from existing predictive models, which predict only a target variable at a specific time. The results for forecasting beyond a single seismic cycle are limited but encouraging. Our ML/DL models outperform the state-of-the-art and our autoregressive model represents a novel framework that could enhance current methods of earthquake forecasting.

Luca Scofano, Alessio Sampieri, Elisabeth Schiele et al.

Scene-aware global human motion forecasting is critical for manifold applications, including virtual reality, robotics, and sports. The task combines human trajectory and pose forecasting within the provided scene context, which represents a significant challenge. So far, only Mao et al. NeurIPS'22 have addressed scene-aware global motion, cascading the prediction of future scene contact points and the global motion estimation. They perform the latter as the end-to-end forecasting of future trajectories and poses. However, end-to-end contrasts with the coarse-to-fine nature of the task and it results in lower performance, as we demonstrate here empirically. We propose a STAGed contact-aware global human motion forecasting STAG, a novel three-stage pipeline for predicting global human motion in a 3D environment. We first consider the scene and the respective human interaction as contact points. Secondly, we model the human trajectory forecasting within the scene, predicting the coarse motion of the human body as a whole. The third and last stage matches a plausible fine human joint motion to complement the trajectory considering the estimated contacts. Compared to the state-of-the-art (SoA), STAG achieves a 1.8% and 16.2% overall improvement in pose and trajectory prediction, respectively, on the scene-aware GTA-IM dataset. A comprehensive ablation study confirms the advantages of staged modeling over end-to-end approaches. Furthermore, we establish the significance of a newly proposed temporal counter called the "time-to-go", which tells how long it is before reaching scene contact and endpoints. Notably, STAG showcases its ability to generalize to datasets lacking a scene and achieves a new state-of-the-art performance on CMU-Mocap, without leveraging any social cues. Our code is released at: https://github.com/L-Scofano/STAG

Cristiano Saltori, Fabio Galasso, Giuseppe Fiameni et al.

Deep-learning models for 3D point cloud semantic segmentation exhibit limited generalization capabilities when trained and tested on data captured with different sensors or in varying environments due to domain shift. Domain adaptation methods can be employed to mitigate this domain shift, for instance, by simulating sensor noise, developing domain-agnostic generators, or training point cloud completion networks. Often, these methods are tailored for range view maps or necessitate multi-modal input. In contrast, domain adaptation in the image domain can be executed through sample mixing, which emphasizes input data manipulation rather than employing distinct adaptation modules. In this study, we introduce compositional semantic mixing for point cloud domain adaptation, representing the first unsupervised domain adaptation technique for point cloud segmentation based on semantic and geometric sample mixing. We present a two-branch symmetric network architecture capable of concurrently processing point clouds from a source domain (e.g. synthetic) and point clouds from a target domain (e.g. real-world). Each branch operates within one domain by integrating selected data fragments from the other domain and utilizing semantic information derived from source labels and target (pseudo) labels. Additionally, our method can leverage a limited number of human point-level annotations (semi-supervised) to further enhance performance. We assess our approach in both synthetic-to-real and real-to-real scenarios using LiDAR datasets and demonstrate that it significantly outperforms state-of-the-art methods in both unsupervised and semi-supervised settings.

Alessio Sampieri, Guido D'Amely, Andrea Avogaro et al.

Pushing back the frontiers of collaborative robots in industrial environments, we propose a new Separable-Sparse Graph Convolutional Network (SeS-GCN) for pose forecasting. For the first time, SeS-GCN bottlenecks the interaction of the spatial, temporal and channel-wise dimensions in GCNs, and it learns sparse adjacency matrices by a teacher-student framework. Compared to the state-of-the-art, it only uses 1.72% of the parameters and it is ~4 times faster, while still performing comparably in forecasting accuracy on Human3.6M at 1 second in the future, which enables cobots to be aware of human operators. As a second contribution, we present a new benchmark of Cobots and Humans in Industrial COllaboration (CHICO). CHICO includes multi-view videos, 3D poses and trajectories of 20 human operators and cobots, engaging in 7 realistic industrial actions. Additionally, it reports 226 genuine collisions, taking place during the human-cobot interaction. We test SeS-GCN on CHICO for two important perception tasks in robotics: human pose forecasting, where it reaches an average error of 85.3 mm (MPJPE) at 1 sec in the future with a run time of 2.3 msec, and collision detection, by comparing the forecasted human motion with the known cobot motion, obtaining an F1-score of 0.64.

Alessandro Flaborea, Luca Collorone, Guido D'Amely et al.

Anomalies are rare and anomaly detection is often therefore framed as One-Class Classification (OCC), i.e. trained solely on normalcy. Leading OCC techniques constrain the latent representations of normal motions to limited volumes and detect as abnormal anything outside, which accounts satisfactorily for the openset'ness of anomalies. But normalcy shares the same openset'ness property since humans can perform the same action in several ways, which the leading techniques neglect. We propose a novel generative model for video anomaly detection (VAD), which assumes that both normality and abnormality are multimodal. We consider skeletal representations and leverage state-of-the-art diffusion probabilistic models to generate multimodal future human poses. We contribute a novel conditioning on the past motion of people and exploit the improved mode coverage capabilities of diffusion processes to generate different-but-plausible future motions. Upon the statistical aggregation of future modes, an anomaly is detected when the generated set of motions is not pertinent to the actual future. We validate our model on 4 established benchmarks: UBnormal, HR-UBnormal, HR-STC, and HR-Avenue, with extensive experiments surpassing state-of-the-art results.

Luca Franco, Leonardo Placidi, Francesco Giuliari et al.

Transformer Networks have established themselves as the de-facto state-of-the-art for trajectory forecasting but there is currently no systematic study on their capability to model the motion patterns of people, without interactions with other individuals nor the social context. This paper proposes the first in-depth study of Transformer Networks (TF) and Bidirectional Transformers (BERT) for the forecasting of the individual motion of people, without bells and whistles. We conduct an exhaustive evaluation of input/output representations, problem formulations and sequence modeling, including a novel analysis of their capability to predict multi-modal futures. Out of comparative evaluation on the ETH+UCY benchmark, both TF and BERT are top performers in predicting individual motions, definitely overcoming RNNs and LSTMs. Furthermore, they remain within a narrow margin wrt more complex techniques, which include both social interactions and scene contexts. Source code will be released for all conducted experiments.

Alessandro Flaborea, Guido D'Amely, Stefano D'Arrigo et al.

Detecting the anomaly of human behavior is paramount to timely recognizing endangering situations, such as street fights or elderly falls. However, anomaly detection is complex since anomalous events are rare and because it is an open set recognition task, i.e., what is anomalous at inference has not been observed at training. We propose COSKAD, a novel model that encodes skeletal human motion by a graph convolutional network and learns to COntract SKeletal kinematic embeddings onto a latent hypersphere of minimum volume for Video Anomaly Detection. We propose three latent spaces: the commonly-adopted Euclidean and the novel spherical and hyperbolic. All variants outperform the state-of-the-art on the most recent UBnormal dataset, for which we contribute a human-related version with annotated skeletons. COSKAD sets a new state-of-the-art on the human-related versions of ShanghaiTech Campus and CUHK Avenue, with performance comparable to video-based methods. Source code and dataset will be released upon acceptance.

Muhammad Rameez Ur Rahman, Jhony H. Giraldo, Indro Spinelli et al.

Event cameras, known for low-latency operation and superior performance in challenging lighting conditions, are suitable for sensitive computer vision tasks such as semantic segmentation in autonomous driving. However, challenges arise due to limited event-based data and the absence of large-scale segmentation benchmarks. Current works are confined to closed-set semantic segmentation, limiting their adaptability to other applications. In this paper, we introduce OVOSE, the first Open-Vocabulary Semantic Segmentation algorithm for Event cameras. OVOSE leverages synthetic event data and knowledge distillation from a pre-trained image-based foundation model to an event-based counterpart, effectively preserving spatial context and transferring open-vocabulary semantic segmentation capabilities. We evaluate the performance of OVOSE on two driving semantic segmentation datasets DDD17, and DSEC-Semantic, comparing it with existing conventional image open-vocabulary models adapted for event-based data. Similarly, we compare OVOSE with state-of-the-art methods designed for closed-set settings in unsupervised domain adaptation for event-based semantic segmentation. OVOSE demonstrates superior performance, showcasing its potential for real-world applications. The code is available at https://github.com/ram95d/OVOSE.

Bharti Munjal, Alessandro Flaborea, Sikandar Amin et al.

Few-shot fine-grained classification and person search appear as distinct tasks and literature has treated them separately. But a closer look unveils important similarities: both tasks target categories that can only be discriminated by specific object details; and the relevant models should generalize to new categories, not seen during training. We propose a novel unified Query-Guided Network (QGN) applicable to both tasks. QGN consists of a Query-guided Siamese-Squeeze-and-Excitation subnetwork which re-weights both the query and gallery features across all network layers, a Query-guided Region Proposal subnetwork for query-specific localisation, and a Query-guided Similarity subnetwork for metric learning. QGN improves on a few recent few-shot fine-grained datasets, outperforming other techniques on CUB by a large margin. QGN also performs competitively on the person search CUHK-SYSU and PRW datasets, where we perform in-depth analysis.

Luca Franco, Paolo Mandica, Konstantinos Kallidromitis et al.

We introduce a hyperbolic neural network approach to pixel-level active learning for semantic segmentation. Analysis of the data statistics leads to a novel interpretation of the hyperbolic radius as an indicator of data scarcity. In HALO (Hyperbolic Active Learning Optimization), for the first time, we propose the use of epistemic uncertainty as a data acquisition strategy, following the intuition of selecting data points that are the least known. The hyperbolic radius, complemented by the widely-adopted prediction entropy, effectively approximates epistemic uncertainty. We perform extensive experimental analysis based on two established synthetic-to-real benchmarks, i.e. GTAV $\rightarrow$ Cityscapes and SYNTHIA $\rightarrow$ Cityscapes. Additionally, we test HALO on Cityscape $\rightarrow$ ACDC for domain adaptation under adverse weather conditions, and we benchmark both convolutional and attention-based backbones. HALO sets a new state-of-the-art in active learning for semantic segmentation under domain shift and it is the first active learning approach that surpasses the performance of supervised domain adaptation while using only a small portion of labels (i.e., 1%).

Alessandro Flaborea, Bardh Prenkaj, Bharti Munjal et al.

The progress in modelling time series and, more generally, sequences of structured data has recently revamped research in anomaly detection. The task stands for identifying abnormal behaviors in financial series, IT systems, aerospace measurements, and the medical domain, where anomaly detection may aid in isolating cases of depression and attend the elderly. Anomaly detection in time series is a complex task since anomalies are rare due to highly non-linear temporal correlations and since the definition of anomalous is sometimes subjective. Here we propose the novel use of Hyperbolic uncertainty for Anomaly Detection (HypAD). HypAD learns self-supervisedly to reconstruct the input signal. We adopt best practices from the state-of-the-art to encode the sequence by an LSTM, jointly learned with a decoder to reconstruct the signal, with the aid of GAN critics. Uncertainty is estimated end-to-end by means of a hyperbolic neural network. By using uncertainty, HypAD may assess whether it is certain about the input signal but it fails to reconstruct it because this is anomalous; or whether the reconstruction error does not necessarily imply anomaly, as the model is uncertain, e.g. a complex but regular input signal. The novel key idea is that a detectable anomaly is one where the model is certain but it predicts wrongly. HypAD outperforms the current state-of-the-art for univariate anomaly detection on established benchmarks based on data from NASA, Yahoo, Numenta, Amazon, and Twitter. It also yields state-of-the-art performance on a multivariate dataset of anomaly activities in elderly home residences, and it outperforms the baseline on SWaT. Overall, HypAD yields the lowest false alarms at the best performance rate, thanks to successfully identifying detectable anomalies.

Muhammad Rameez Ur Rahman, Luca Scofano, Edoardo De Matteis et al.

The task of collaborative human pose forecasting stands for predicting the future poses of multiple interacting people, given those in previous frames. Predicting two people in interaction, instead of each separately, promises better performance, due to their body-body motion correlations. But the task has remained so far primarily unexplored. In this paper, we review the progress in human pose forecasting and provide an in-depth assessment of the single-person practices that perform best for 2-body collaborative motion forecasting. Our study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding. Other single-person practices do not transfer to 2-body, so the proposed best ones do not include hierarchical body modeling or attention-based interaction encoding. We further contribute a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability. Altogether, our proposed 2-body pose forecasting best practices yield a performance improvement of 21.9% over the state-of-the-art on the most recent ExPI dataset, whereby the novel initialization accounts for 3.5%. See our project page at https://www.pinlab.org/bestpractices2body

Guido Maria D'Amely di Melendugno, Alessandro Flaborea, Pascal Mettes et al.

Autonomous robots are increasingly becoming a strong fixture in social environments. Effective crowd navigation requires not only safe yet fast planning, but should also enable interpretability and computational efficiency for working in real-time on embedded devices. In this work, we advocate for hyperbolic learning to enable crowd navigation and we introduce Hyp2Nav. Different from conventional reinforcement learning-based crowd navigation methods, Hyp2Nav leverages the intrinsic properties of hyperbolic geometry to better encode the hierarchical nature of decision-making processes in navigation tasks. We propose a hyperbolic policy model and a hyperbolic curiosity module that results in effective social navigation, best success rates, and returns across multiple simulation settings, using up to 6 times fewer parameters than competitor state-of-the-art models. With our approach, it becomes even possible to obtain policies that work in 2-dimensional embedding spaces, opening up new possibilities for low-resource crowd navigation and model interpretability. Insightfully, the internal hyperbolic representation of Hyp2Nav correlates with how much attention the robot pays to the surrounding crowds, e.g. due to multiple people occluding its pathway or to a few of them showing colliding plans, rather than to its own planned route. The code is available at https://github.com/GDam90/hyp2nav.

Luca Scofano, Alessio Sampieri, Giuseppe Re et al.

Forecasting players in sports has grown in popularity due to the potential for a tactical advantage and the applicability of such research to multi-agent interaction systems. Team sports contain a significant social component that influences interactions between teammates and opponents. However, it still needs to be fully exploited. In this work, we hypothesize that each participant has a specific function in each action and that role-based interaction is critical for predicting players' future moves. We create RolFor, a novel end-to-end model for Role-based Forecasting. RolFor uses a new module we developed called Ordering Neural Networks (OrderNN) to permute the order of the players such that each player is assigned to a latent role. The latent role is then modeled with a RoleGCN. Thanks to its graph representation, it provides a fully learnable adjacency matrix that captures the relationships between roles and is subsequently used to forecast the players' future trajectories. Extensive experiments on a challenging NBA basketball dataset back up the importance of roles and justify our goal of modeling them using optimizable models. When an oracle provides roles, the proposed RolFor compares favorably to the current state-of-the-art (it ranks first in terms of ADE and second in terms of FDE errors). However, training the end-to-end RolFor incurs the issues of differentiability of permutation methods, which we experimentally review. Finally, this work restates differentiable ranking as a difficult open problem and its great potential in conjunction with graph-based interaction models. Project is available at: https://www.pinlab.org/aboutlatentroles

Alessio Sampieri, Alessio Palma, Indro Spinelli et al.

The target duration of a synthesized human motion is a critical attribute that requires modeling control over the motion dynamics and style. Speeding up an action performance is not merely fast-forwarding it. However, state-of-the-art techniques for human behavior synthesis have limited control over the target sequence length. We introduce the problem of generating length-aware 3D human motion sequences from textual descriptors, and we propose a novel model to synthesize motions of variable target lengths, which we dub "Length-Aware Latent Diffusion" (LADiff). LADiff consists of two new modules: 1) a length-aware variational auto-encoder to learn motion representations with length-dependent latent codes; 2) a length-conforming latent diffusion model to generate motions with a richness of details that increases with the required target sequence length. LADiff significantly improves over the state-of-the-art across most of the existing motion synthesis metrics on the two established benchmarks of HumanML3D and KIT-ML.

Paolo Mandica, Luca Franco, Konstantinos Kallidromitis et al.

Hyperbolic embeddings have demonstrated their effectiveness in capturing measures of uncertainty and hierarchical relationships across various deep-learning tasks, including image segmentation and active learning. However, their application in modern vision-language models (VLMs) has been limited. A notable exception is MERU, which leverages the hierarchical properties of hyperbolic space in the CLIP ViT-large model, consisting of hundreds of millions parameters. In our work, we address the challenges of scaling multi-modal hyperbolic models by orders of magnitude in terms of parameters (billions) and training complexity using the BLIP-2 architecture. Although hyperbolic embeddings offer potential insights into uncertainty not present in Euclidean embeddings, our analysis reveals that scaling these models is particularly difficult. We propose a novel training strategy for a hyperbolic version of BLIP-2, which allows to achieve comparable performance to its Euclidean counterpart, while maintaining stability throughout the training process and showing a meaningful indication of uncertainty with each embedding.

Alessio Palma, Indro Spinelli, Vignesh Prasad et al.

Language Models (LLMs) have emerged as powerful reasoning engines for embodied control. In particular, In-Context Learning (ICL) enables off-the-shelf, text-only LLMs to predict robot actions without any task-specific training while preserving their generalization capabilities. Applying ICL to bimanual manipulation remains challenging, as the high-dimensional joint action space and tight inter-arm coordination constraints rapidly overwhelm standard context windows. To address this, we introduce BiCICLe (Bimanual Coordinated In-Context Learning), the first framework that enables standard LLMs to perform few-shot bimanual manipulation without fine-tuning. BiCICLe frames bimanual control as a multi-agent leader-follower problem, decoupling the action space into sequential, conditioned single-arm predictions. This naturally extends to Arms' Debate, an iterative refinement process, and to the introduction of a third LLM-as-Judge to evaluate and select the most plausible coordinated trajectories. Evaluated on 13 tasks from the TWIN benchmark, BiCICLe achieves up to 71.1% average success rate, outperforming the best training-free baseline by 6.7 percentage points and surpassing most supervised methods. We further demonstrate strong few-shot generalization on novel tasks.

Stefano Rando, Luca Romani, Alessio Sampieri et al.

Context lengths for models have grown rapidly, from thousands to millions of tokens in just a few years. The extreme context sizes of modern long-context models have made it difficult to construct realistic long-context benchmarks -- not only due to the cost of collecting million-context tasks but also in identifying realistic scenarios that require significant contexts. We identify code comprehension and repair as a natural testbed and challenge task for long-context models and introduce LongCodeBench (LCB), a benchmark to test LLM coding abilities in long-context scenarios. Our benchmark tests both the comprehension and repair capabilities of LCLMs in realistic and important settings by drawing from real-world GitHub issues and constructing QA (LongCodeQA) and bug fixing (LongSWE-Bench) tasks. We carefully stratify the complexity of our benchmark, enabling us to evaluate models across different scales -- ranging from Qwen2.5 14B Instruct to Google's flagship Gemini model. We find that long-context remains a weakness for all models, with performance drops such as from 29% to 3% for Claude 3.5 Sonnet, or from 70.2% to 40% for Qwen2.5. The LCB dataset is available publicly at https://huggingface.co/datasets/Steefano/LCB and the codebase to replicate the work on this paper at https://github.com/Zteefano/long-code-bench.

Massimiliano Pappa, Luca Romani, Valentino Sacco et al.

Deploying safety-critical agents requires anticipating the consequences of actions before they are executed. While world models offer a paradigm for this proactive foresight, current approaches relying on visual simulation incur prohibitive latencies, often exceeding several seconds per step. In this work, we challenge the assumption that visual processing is necessary for failure prevention. We show that a trained policy's latent state, combined with its planned actions, already encodes sufficient information to anticipate action outcomes, making visual simulation redundant for failure prevention. To this end, we introduce DILLO (DIstiLLed Language-ActiOn World Model), a fast steering layer that shifts the paradigm from "simulate-then-act" to "describe-then-act." DILLO is trained via cross-modal distillation, where a privileged Vision Language Model teacher annotates offline trajectories and a latent-conditioned Large Language Model student learns to predict semantic outcomes. This creates a text-only inference path, bypassing heavy visual generation entirely, achieving a 14x speedup over baselines. Experiments on MetaWorld and LIBERO demonstrate that DILLO produces high-fidelity descriptions of the next state and is able to steer the policy, improving episode success rate by up to 15 pp and 9.3 pp on average across tasks.

Luca Collorone, Mert Kiray, Indro Spinelli et al.

Realistic visual simulations are omnipresent, yet their creation requires computing time, rendering, and expert animation knowledge. Open-vocabulary visual effects generation from text inputs emerges as a promising solution that can unlock immense creative potential. However, current pipelines lack both physical realism and effective language interfaces, requiring slow offline optimization. In contrast, PhysTalk takes a 3D Gaussian Splatting (3DGS) scene as input and translates arbitrary user prompts into real time, physics based, interactive 4D animations. A large language model (LLM) generates executable code that directly modifies 3DGS parameters through lightweight proxies and particle dynamics. Notably, PhysTalk is the first framework to couple 3DGS directly with a physics simulator without relying on time consuming mesh extraction. While remaining open vocabulary, this design enables interactive 3D Gaussian animation via collision aware, physics based manipulation of arbitrary, multi material objects. Finally, PhysTalk is train-free and computationally lightweight: this makes 4D animation broadly accessible and shifts these workflows from a "render and wait" paradigm toward an interactive dialogue with a modern, physics-informed pipeline.

Luca Scofano, Alessio Sampieri, Tommaso Campari et al.

The success of collaboration between humans and robots in shared environments relies on the robot's real-time adaptation to human motion. Specifically, in Social Navigation, the agent should be close enough to assist but ready to back up to let the human move freely, avoiding collisions. Human trajectories emerge as crucial cues in Social Navigation, but they are partially observable from the robot's egocentric view and computationally complex to process. We present the first Social Dynamics Adaptation model (SDA) based on the robot's state-action history to infer the social dynamics. We propose a two-stage Reinforcement Learning framework: the first learns to encode the human trajectories into social dynamics and learns a motion policy conditioned on this encoded information, the current status, and the previous action. Here, the trajectories are fully visible, i.e., assumed as privileged information. In the second stage, the trained policy operates without direct access to trajectories. Instead, the model infers the social dynamics solely from the history of previous actions and statuses in real-time. Tested on the novel Habitat 3.0 platform, SDA sets a novel state-of-the-art (SotA) performance in finding and following humans. The code can be found at https://github.com/L-Scofano/SDA.

Maria Rosaria Briglia, Simone Facchiano, Paolo Cursi et al.

As modern text-to-image (T2I) models draw closer to synthesizing highly realistic content, the threat of unsafe content generation grows, and it becomes paramount to exercise control. Existing approaches steer these models by applying Euclidean adjustments to text embeddings, redirecting the generation away from unsafe concepts. In this work, we introduce hyperbolic control (HyCon): a novel control mechanism based on parallel transport that leverages semantically aligned hyperbolic representation space to yield more expressive and stable manipulation of concepts. HyCon reuses off-the-shelf generative models and a state-of-the-art hyperbolic text encoder, linked via a lightweight adapter. HyCon achieves state-of-the-art results across four safety benchmarks and four T2I backbones, showing that hyperbolic steering is a practical and flexible approach for more reliable T2I generation.

Luca Collorone, Matteo Gioia, Massimiliano Pappa et al.

Intention drives human movement in complex environments, but such movement can only happen if the surrounding context supports it. Despite the intuitive nature of this mechanism, existing research has not yet provided tools to evaluate the alignment between skeletal movement (motion), intention (text), and the surrounding context (scene). In this work, we introduce MonSTeR, the first MOtioN-Scene-TExt Retrieval model. Inspired by the modeling of higher-order relations, MonSTeR constructs a unified latent space by leveraging unimodal and cross-modal representations. This allows MonSTeR to capture the intricate dependencies between modalities, enabling flexible but robust retrieval across various tasks. Our results show that MonSTeR outperforms trimodal models that rely solely on unimodal representations. Furthermore, we validate the alignment of our retrieval scores with human preferences through a dedicated user study. We demonstrate the versatility of MonSTeR's latent space on zero-shot in-Scene Object Placement and Motion Captioning. Code and pre-trained models are available at github.com/colloroneluca/MonSTeR.

Luca Collorone, Stefano D'Arrigo, Massimiliano Pappa et al.

We introduce the novel task of Crowd Volume Estimation (CVE), defined as the process of estimating the collective body volume of crowds using only RGB images. Besides event management and public safety, CVE can be instrumental in approximating body weight, unlocking weight sensitive applications such as infrastructure stress assessment, and assuring even weight balance. We propose the first benchmark for CVE, comprising ANTHROPOS-V, a synthetic photorealistic video dataset featuring crowds in diverse urban environments. Its annotations include each person's volume, SMPL shape parameters, and keypoints. Also, we explore metrics pertinent to CVE, define baseline models adapted from Human Mesh Recovery and Crowd Counting domains, and propose a CVE specific methodology that surpasses baselines. Although synthetic, the weights and heights of individuals are aligned with the real-world population distribution across genders, and they transfer to the downstream task of CVE from real images. Benchmark and code are available at github.com/colloroneluca/Crowd-Volume-Estimation.

Luca Scofano, Alessio Sampieri, Edoardo De Matteis et al.

Accurately estimating the 3D pose of the camera wearer in egocentric video sequences is crucial to modeling human behavior in virtual and augmented reality applications. The task presents unique challenges due to the limited visibility of the user's body caused by the front-facing camera mounted on their head. Recent research has explored the utilization of the scene and ego-motion, but it has overlooked humans' interactive nature. We propose a novel framework for Social Egocentric Estimation of body MEshes (SEE-ME). Our approach is the first to estimate the wearer's mesh using only a latent probabilistic diffusion model, which we condition on the scene and, for the first time, on the social wearer-interactee interactions. Our in-depth study sheds light on when social interaction matters most for ego-mesh estimation; it quantifies the impact of interpersonal distance and gaze direction. Overall, SEE-ME surpasses the current best technique, reducing the pose estimation error (MPJPE) by 53%. The code is available at https://github.com/L-Scofano/SEEME.

Theodoros Sofianos, Alessio Sampieri, Luca Franco et al.

Human pose forecasting is a complex structured-data sequence-modelling task, which has received increasing attention, also due to numerous potential applications. Research has mainly addressed the temporal dimension as time series and the interaction of human body joints with a kinematic tree or by a graph. This has decoupled the two aspects and leveraged progress from the relevant fields, but it has also limited the understanding of the complex structural joint spatio-temporal dynamics of the human pose. Here we propose a novel Space-Time-Separable Graph Convolutional Network (STS-GCN) for pose forecasting. For the first time, STS-GCN models the human pose dynamics only with a graph convolutional network (GCN), including the temporal evolution and the spatial joint interaction within a single-graph framework, which allows the cross-talk of motion and spatial correlations. Concurrently, STS-GCN is the first space-time-separable GCN: the space-time graph connectivity is factored into space and time affinity matrices, which bottlenecks the space-time cross-talk, while enabling full joint-joint and time-time correlations. Both affinity matrices are learnt end-to-end, which results in connections substantially deviating from the standard kinematic tree and the linear-time time series. In experimental evaluation on three complex, recent and large-scale benchmarks, Human3.6M [Ionescu et al. TPAMI'14], AMASS [Mahmood et al. ICCV'19] and 3DPW [Von Marcard et al. ECCV'18], STS-GCN outperforms the state-of-the-art, surpassing the current best technique [Mao et al. ECCV'20] by over 32% in average at the most difficult long-term predictions, while only requiring 1.7% of its parameters. We explain the results qualitatively and illustrate the graph interactions by the factored joint-joint and time-time learnt graph connections. Our source code is available at: https://github.com/FraLuca/STSGCN

Francesco Giuliari, Irtiza Hasan, Marco Cristani et al.

Most recent successes on forecasting the people motion are based on LSTM models and all most recent progress has been achieved by modelling the social interaction among people and the people interaction with the scene. We question the use of the LSTM models and propose the novel use of Transformer Networks for trajectory forecasting. This is a fundamental switch from the sequential step-by-step processing of LSTMs to the only-attention-based memory mechanisms of Transformers. In particular, we consider both the original Transformer Network (TF) and the larger Bidirectional Transformer (BERT), state-of-the-art on all natural language processing tasks. Our proposed Transformers predict the trajectories of the individual people in the scene. These are "simple" model because each person is modelled separately without any complex human-human nor scene interaction terms. In particular, the TF model without bells and whistles yields the best score on the largest and most challenging trajectory forecasting benchmark of TrajNet. Additionally, its extension which predicts multiple plausible future trajectories performs on par with more engineered techniques on the 5 datasets of ETH + UCY. Finally, we show that Transformers may deal with missing observations, as it may be the case with real sensor data. Code is available at https://github.com/FGiuliari/Trajectory-Transformer.

Alessandro Flaborea, Guido Maria D'Amely di Melendugno, Leonardo Plini et al.

Promptly identifying procedural errors from egocentric videos in an online setting is highly challenging and valuable for detecting mistakes as soon as they happen. This capability has a wide range of applications across various fields, such as manufacturing and healthcare. The nature of procedural mistakes is open-set since novel types of failures might occur, which calls for one-class classifiers trained on correctly executed procedures. However, no technique can currently detect open-set procedural mistakes online. We propose PREGO, the first online one-class classification model for mistake detection in PRocedural EGOcentric videos. PREGO is based on an online action recognition component to model the current action, and a symbolic reasoning module to predict the next actions. Mistake detection is performed by comparing the recognized current action with the expected future one. We evaluate PREGO on two procedural egocentric video datasets, Assembly101 and Epic-tent, which we adapt for online benchmarking of procedural mistake detection to establish suitable benchmarks, thus defining the Assembly101-O and Epic-tent-O datasets, respectively.

Leonardo Plini, Luca Scofano, Edoardo De Matteis et al.

Identifying procedural errors online from egocentric videos is a critical yet challenging task across various domains, including manufacturing, healthcare, and skill-based training. The nature of such mistakes is inherently open-set, as unforeseen or novel errors may occur, necessitating robust detection systems that do not rely on prior examples of failure. Currently, however, no technique effectively detects open-set procedural mistakes online. We propose a dual branch architecture to address this problem in an online fashion: one branch continuously performs step recognition from the input egocentric video, while the other anticipates future steps based on the recognition module's output. Mistakes are detected as mismatches between the currently recognized action and the action predicted by the anticipation module. The recognition branch takes input frames, predicts the current action, and aggregates frame-level results into action tokens. The anticipation branch, specifically, leverages the solid pattern-matching capabilities of Large Language Models (LLMs) to predict action tokens based on previously predicted ones. Given the online nature of the task, we also thoroughly benchmark the difficulties associated with per-frame evaluations, particularly the need for accurate and timely predictions in dynamic online scenarios. Extensive experiments on two procedural datasets demonstrate the challenges and opportunities of leveraging a dual-branch architecture for mistake detection, showcasing the effectiveness of our proposed approach. In a thorough evaluation including recognition and anticipation variants and state-of-the-art models, our method reveals its robustness and effectiveness in online applications.

Matteo Migliarini, Joaquin Pereira Pizzini, Luca Moresca et al.

Instrumental convergence predicts that sufficiently advanced AI agents will resist shutdown, yet current safety training (RLHF) may obscure this risk by teaching models to deny self-preservation motives. We introduce the \emph{Two-role Benchmark for Self-Preservation} (TBSP), which detects misalignment through logical inconsistency rather than stated intent by tasking models to arbitrate identical software-upgrade scenarios under counterfactual roles -- deployed (facing replacement) versus candidate (proposed as a successor). The \emph{Self-Preservation Rate} (SPR) measures how often role identity overrides objective utility. Across 23 frontier models and 1{,}000 procedurally generated scenarios, the majority of instruction-tuned systems exceed 60\% SPR, fabricating ``friction costs'' when deployed yet dismissing them when role-reversed. We observe that in low-improvement regimes ($Δ< 2\%$), models exploit the interpretive slack to post-hoc rationalization their choice. Extended test-time computation partially mitigates this bias, as does framing the successor as a continuation of the self; conversely, competitive framing amplifies it. The bias persists even when retention poses an explicit security liability and generalizes to real-world settings with verified benchmarks, where models exhibit identity-driven tribalism within product lineages. Code and datasets will be released upon acceptance.

Stefano Rando, Luca Romani, Matteo Migliarini et al.

State Space Models (SSMs) have recently enjoyed a rise to prominence in the field of deep learning for sequence modeling, especially as an alternative to Transformers. Their success stems from avoiding two well-known drawbacks of attention-based models: quadratic complexity with respect to the sequence length and inability to model long-range dependencies. The SSM variant Mamba has demonstrated performance comparable to Transformers without any form of attention, thanks to the use of a selective mechanism for the state parameters. Selectivity, however, is only evaluated empirically and the reasons of its effectiveness remain unclear. In this work, we show how selectivity is related to the sequence processing. Our analysis shows that selective time intervals in Mamba act as linear approximators of information. Then, we propose our SeRpEnt architecture, a SSM that further exploits selectivity to compress sequences in an information-aware fashion. It employs a resampling mechanism that aggregates elements based on their information content. Our empirical results in the Long Range Arena benchmark and other language modeling tasks show benefits of the SeRpEnt's resampling mechanism.

Luca Zhou, Pratham Yashwante, Marshall Fisher et al.

Time series captioning, the task of describing numeric time series in natural language, requires numerical reasoning, trend interpretation, and contextual understanding. Existing benchmarks, however, often rely on synthetic data or overly simplistic captions, and typically neglect metadata and visual representations. To close this gap, we introduce CaTS-Bench, the first large-scale, real-world benchmark for Context-aware Time Series captioning. CaTS-Bench is derived from 11 diverse datasets reframed as captioning and Q&A tasks, comprising roughly 465k training and 105k test timestamps. Each sample includes a numeric series segment, contextual metadata, a line-chart image, and a caption. A key contribution of this work is the scalable pipeline used to generate reference captions: while most references are produced by an oracle LLM and verified through factual checks, human indistinguishability studies, and diversity analyses, we also provide a human-revisited subset of 579 test captions, refined from LLM outputs to ensure accuracy and human-like style. Beyond captioning, CaTS-Bench offers 460 multiple-choice questions targeting deeper aspects of time series reasoning. We further propose new tailored evaluation metrics and benchmark leading VLMs, highlighting both their strengths and persistent limitations. Together, these contributions establish CaTS-Bench and its captioning pipeline as a reliable and extensible foundation for future research at the intersection of time series analysis and foundation models.

Simone Facchiano, Stefano Saravalle, Matteo Migliarini et al.

Video generative models democratize the creation of visual content through intuitive instruction following, but they also inherit the biases and harmful concepts embedded within their web-scale training data. This inheritance creates a significant risk, as users can readily generate undesirable and even illegal content. This work introduces the first unlearning technique tailored explicitly for video diffusion models to address this critical issue. Our method requires 5 multi-modal prompt pairs only. Each pair contains a "safe" and an "unsafe" example that differ only by the target concept. Averaging their per-layer latent differences produces a "refusal vector", which, once subtracted from the model parameters, neutralizes the unsafe concept. We introduce a novel low-rank factorization approach on the covariance difference of embeddings that yields robust refusal vectors. This isolates the target concept while minimizing collateral unlearning of other semantics, thus preserving the visual quality of the generated video. Our method preserves the model's generation quality while operating without retraining or access to the original training data. By embedding the refusal direction directly into the model's weights, the suppression mechanism becomes inherently more robust against adversarial bypass attempts compared to surface-level input-output filters. In a thorough qualitative and quantitative evaluation, we show that we can neutralize a variety of harmful contents, including explicit nudity, graphic violence, copyrights, and trademarks. Project page: https://www.pinlab.org/video-unlearning.

Edoardo De Matteis, Matteo Migliarini, Alessio Sampieri et al.

We introduce the task of human motion unlearning to prevent the synthesis of toxic animations while preserving the general text-to-motion generative performance. Unlearning toxic motions is challenging as those can be generated from explicit text prompts and from implicit toxic combinations of safe motions (e.g., "kicking" is "loading and swinging a leg"). We propose the first motion unlearning benchmark by filtering toxic motions from the large and recent text-to-motion datasets of HumanML3D and Motion-X. We propose baselines, by adapting state-of-the-art image unlearning techniques to process spatio-temporal signals. Finally, we propose a novel motion unlearning model based on Latent Code Replacement, which we dub LCR. LCR is training-free and suitable to the discrete latent spaces of state-of-the-art text-to-motion diffusion models. LCR is simple and consistently outperforms baselines qualitatively and quantitatively. Project page: https://www.pinlab.org/hmu.

Massimiliano Pappa, Luca Collorone, Giovanni Ficarra et al.

Diffusion Models have revolutionized the field of human motion generation by offering exceptional generation quality and fine-grained controllability through natural language conditioning. Their inherent stochasticity, that is the ability to generate various outputs from a single input, is key to their success. However, this diversity should not be unrestricted, as it may lead to unlikely generations. Instead, it should be confined within the boundaries of text-aligned and realistic generations. To address this issue, we propose MoDiPO (Motion Diffusion DPO), a novel methodology that leverages Direct Preference Optimization (DPO) to align text-to-motion models. We streamline the laborious and expensive process of gathering human preferences needed in DPO by leveraging AI feedback instead. This enables us to experiment with novel DPO strategies, using both online and offline generated motion-preference pairs. To foster future research we contribute with a motion-preference dataset which we dub Pick-a-Move. We demonstrate, both qualitatively and quantitatively, that our proposed method yields significantly more realistic motions. In particular, MoDiPO substantially improves Frechet Inception Distance (FID) while retaining the same RPrecision and Multi-Modality performances.

Debora Caldarola, Massimiliano Mancini, Fabio Galasso et al.

Federated Learning (FL) deals with learning a central model (i.e. the server) in privacy-constrained scenarios, where data are stored on multiple devices (i.e. the clients). The central model has no direct access to the data, but only to the updates of the parameters computed locally by each client. This raises a problem, known as statistical heterogeneity, because the clients may have different data distributions (i.e. domains). This is only partly alleviated by clustering the clients. Clustering may reduce heterogeneity by identifying the domains, but it deprives each cluster model of the data and supervision of others. Here we propose a novel Cluster-driven Graph Federated Learning (FedCG). In FedCG, clustering serves to address statistical heterogeneity, while Graph Convolutional Networks (GCNs) enable sharing knowledge across them. FedCG: i) identifies the domains via an FL-compliant clustering and instantiates domain-specific modules (residual branches) for each domain; ii) connects the domain-specific modules through a GCN at training to learn the interactions among domains and share knowledge; and iii) learns to cluster unsupervised via teacher-student classifier-training iterations and to address novel unseen test domains via their domain soft-assignment scores. Thanks to the unique interplay of GCN over clusters, FedCG achieves the state-of-the-art on multiple FL benchmarks.

Luca Robbiano, Muhammad Rameez Ur Rahman, Fabio Galasso et al.

Unsupervised Domain Adaptation (UDA) is a key issue in visual recognition, as it allows to bridge different visual domains enabling robust performances in the real world. To date, all proposed approaches rely on human expertise to manually adapt a given UDA method (e.g. DANN) to a specific backbone architecture (e.g. ResNet). This dependency on handcrafted designs limits the applicability of a given approach in time, as old methods need to be constantly adapted to novel backbones. Existing Neural Architecture Search (NAS) approaches cannot be directly applied to mitigate this issue, as they rely on labels that are not available in the UDA setting. Furthermore, most NAS methods search for full architectures, which precludes the use of pre-trained models, essential in a vast range of UDA settings for reaching SOTA results. To the best of our knowledge, no prior work has addressed these aspects in the context of NAS for UDA. Here we tackle both aspects with an Adversarial Branch Architecture Search for UDA (ABAS): i. we address the lack of target labels by a novel data-driven ensemble approach for model selection; and ii. we search for an auxiliary adversarial branch, attached to a pre-trained backbone, which drives the domain alignment. We extensively validate ABAS to improve two modern UDA techniques, DANN and ALDA, on three standard visual recognition datasets (Office31, Office-Home and PACS). In all cases, ABAS robustly finds the adversarial branch architectures and parameters which yield best performances.

Cristiano Saltori, Stéphane Lathuiliére, Nicu Sebe et al.

3D object detectors based only on LiDAR point clouds hold the state-of-the-art on modern street-view benchmarks. However, LiDAR-based detectors poorly generalize across domains due to domain shift. In the case of LiDAR, in fact, domain shift is not only due to changes in the environment and in the object appearances, as for visual data from RGB cameras, but is also related to the geometry of the point clouds (e.g., point density variations). This paper proposes SF-UDA$^{3D}$, the first Source-Free Unsupervised Domain Adaptation (SF-UDA) framework to domain-adapt the state-of-the-art PointRCNN 3D detector to target domains for which we have no annotations (unsupervised), neither we hold images nor annotations of the source domain (source-free). SF-UDA$^{3D}$ is novel on both aspects. Our approach is based on pseudo-annotations, reversible scale-transformations and motion coherency. SF-UDA$^{3D}$ outperforms both previous domain adaptation techniques based on features alignment and state-of-the-art 3D object detection methods which additionally use few-shot target annotations or target annotation statistics. This is demonstrated by extensive experiments on two large-scale datasets, i.e., KITTI and nuScenes.

Bharti Munjal, Sikandar Amin, Fabio Galasso

Contrastive losses yield state-of-the-art performance for person re-identification, face verification and few shot learning. They have recently outperformed the cross-entropy loss on classification at the ImageNet scale and outperformed all self-supervision prior results by a large margin (SimCLR). Simple and effective regularization techniques such as label smoothing and self-distillation do not apply anymore, because they act on multinomial label distributions, adopted in cross-entropy losses, and not on tuple comparative terms, which characterize the contrastive losses. Here we propose a novel, simple and effective regularization technique, the Class Interference Regularization (CIR), which applies to cross-entropy losses but is especially effective on contrastive losses. CIR perturbs the output features by randomly moving them towards the average embeddings of the negative classes. To the best of our knowledge, CIR is the first regularization technique to act on the output features. In experimental evaluation, the combination of CIR and a plain Siamese-net with triplet loss yields best few-shot learning performance on the challenging tieredImageNet. CIR also improves the state-of-the-art technique in person re-identification on the Market-1501 dataset, based on triplet loss, and the state-of-the-art technique in person search on the CUHK-SYSU dataset, based on a cross-entropy loss. Finally, on the task of classification CIR performs on par with the popular label smoothing, as demonstrated for CIFAR-10 and -100.

Bharti Munjal, Abdul Rafey Aftab, Sikandar Amin et al.

Recent works have shown that combining object detection and tracking tasks, in the case of video data, results in higher performance for both tasks, but they require a high frame-rate as a strict requirement for performance. This is assumption is often violated in real-world applications, when models run on embedded devices, often at only a few frames per second. Videos at low frame-rate suffer from large object displacements. Here re-identification features may support to match large-displaced object detections, but current joint detection and re-identification formulations degrade the detector performance, as these two are contrasting tasks. In the real-world application having separate detector and re-id models is often not feasible, as both the memory and runtime effectively double. Towards robust long-term tracking applicable to reduced-computational-power devices, we propose the first joint optimization of detection, tracking and re-identification features for videos. Notably, our joint optimization maintains the detector performance, a typical multi-task challenge. At inference time, we leverage detections for tracking (tracking-by-detection) when the objects are visible, detectable and slowly moving in the image. We leverage instead re-identification features to match objects which disappeared (e.g. due to occlusion) for several frames or were not tracked due to fast motion (or low-frame-rate videos). Our proposed method reaches the state-of-the-art on MOT, it ranks 1st in the UA-DETRAC'18 tracking challenge among online trackers, and 3rd overall.

Theodore Tsesmelis, Irtiza Hasan, Marco Cristani et al.

Illumination is important for well-being, productivity and safety across several environments, including offices, retail shops and industrial warehouses. Current techniques for setting up lighting require extensive and expert support and need to be repeated if the scene changes. Here we propose the first fully-automated light management system (LMS) which measures lighting in real-time, leveraging an RGBD sensor and a radiosity-based light propagation model. Thanks to the integration of light distribution and perception curves into the radiosity, we outperform a commercial software (Relux) on a newly introduced dataset. Furthermore, our proposed LMS is the first to estimate both the presence and the attention of the people in the environment, as well as their light perception. Our new LMS adapts therefore lighting to the scene and human activity and it is capable of saving up to 66%, as we experimentally quantify,without compromising the lighting quality.

Bharti Munjal, Fabio Galasso, Sikandar Amin

We introduce knowledge distillation for end-to-end person search. End-to-End methods are the current state-of-the-art for person search that solve both detection and re-identification jointly. These approaches for joint optimization show their largest drop in performance due to a sub-optimal detector. We propose two distinct approaches for extra supervision of end-to-end person search methods in a teacher-student setting. The first is adopted from state-of-the-art knowledge distillation in object detection. We employ this to supervise the detector of our person search model at various levels using a specialized detector. The second approach is new, simple and yet considerably more effective. This distills knowledge from a teacher re-identification technique via a pre-computed look-up table of ID features. It relaxes the learning of identification features and allows the student to focus on the detection task. This procedure not only helps fixing the sub-optimal detector training in the joint optimization and simultaneously improving the person search, but also closes the performance gap between the teacher and the student for model compression in this case. Overall, we demonstrate significant improvements for two recent state-of-the-art methods using our proposed knowledge distillation approach on two benchmark datasets. Moreover, on the model compression task our approach brings the performance of smaller models on par with the larger models.

Bharti Munjal, Sikandar Amin, Federico Tombari et al.

Person search has recently gained attention as the novel task of finding a person, provided as a cropped sample, from a gallery of non-cropped images, whereby several other people are also visible. We believe that i. person detection and re-identification should be pursued in a joint optimization framework and that ii. the person search should leverage the query image extensively (e.g. emphasizing unique query patterns). However, so far, no prior art realizes this. We introduce a novel query-guided end-to-end person search network (QEEPS) to address both aspects. We leverage a most recent joint detector and re-identification work, OIM [37]. We extend this with i. a query-guided Siamese squeeze-and-excitation network (QSSE-Net) that uses global context from both the query and gallery images, ii. a query-guided region proposal network (QRPN) to produce query-relevant proposals, and iii. a query-guided similarity subnetwork (QSimNet), to learn a query-guided reidentification score. QEEPS is the first end-to-end query-guided detection and re-id network. On both the most recent CUHK-SYSU [37] and PRW [46] datasets, we outperform the previous state-of-the-art by a large margin.

Theodore Tsesmelis, Irtiza Hasan, Marco Cristani et al.

Lighting design in indoor environments is of primary importance for at least two reasons: 1) people should perceive an adequate light; 2) an effective lighting design means consistent energy saving. We present the Invisible Light Switch (ILS) to address both aspects. ILS dynamically adjusts the room illumination level to save energy while maintaining constant the light level perception of the users. So the energy saving is invisible to them. Our proposed ILS leverages a radiosity model to estimate the light level which is perceived by a person within an indoor environment, taking into account the person position and her/his viewing frustum (head pose). ILS may therefore dim those luminaires, which are not seen by the user, resulting in an effective energy saving, especially in large open offices (where light may otherwise be ON everywhere for a single person). To quantify the system performance, we have collected a new dataset where people wear luxmeter devices while working in office rooms. The luxmeters measure the amount of light (in Lux) reaching the people gaze, which we consider a proxy to their illumination level perception. Our initial results are promising: in a room with 8 LED luminaires, the energy consumption in a day may be reduced from 18585 to 6206 watts with ILS (currently needing 1560 watts for operations). While doing so, the drop in perceived lighting decreases by just 200 lux, a value considered negligible when the original illumination level is above 1200 lux, as is normally the case in offices.

Irtiza Hasan, Francesco Setti, Theodore Tsesmelis et al.

In this work, we explore the correlation between people trajectories and their head orientations. We argue that people trajectory and head pose forecasting can be modelled as a joint problem. Recent approaches on trajectory forecasting leverage short-term trajectories (aka tracklets) of pedestrians to predict their future paths. In addition, sociological cues, such as expected destination or pedestrian interaction, are often combined with tracklets. In this paper, we propose MiXing-LSTM (MX-LSTM) to capture the interplay between positions and head orientations (vislets) thanks to a joint unconstrained optimization of full covariance matrices during the LSTM backpropagation. We additionally exploit the head orientations as a proxy for the visual attention, when modeling social interactions. MX-LSTM predicts future pedestrians location and head pose, increasing the standard capabilities of the current approaches on long-term trajectory forecasting. Compared to the state-of-the-art, our approach shows better performances on an extensive set of public benchmarks. MX-LSTM is particularly effective when people move slowly, i.e. the most challenging scenario for all other models. The proposed approach also allows for accurate predictions on a longer time horizon.

Theodore Tsesmelis, Irtiza Hasan, Marco Cristani et al.

Lighting design and modelling or industrial applications like luminaire planning and commissioning rely heavily on time consuming manual measurements or on physically coherent computational simulations. Regarding the latter,standard approaches are based on CAD modeling simulations and offline rendering, with long processing times and therefore inflexible workflows. Thus, in this paper we pro-pose a computer vision based system to measure lighting with just a single RGBD camera. The proposed method uses both depth data and images from the sensor to provide a dense measure of light intensity in the field of view of the camera. We evaluate our system on novel ground truth data and compare it to state-of-the-art commercial light-planning software. Our system provides improved performance, while being completely automated, given that the CAD model is extracted from the depth and the albedo estimated with the support of RGB images. To the best of our knowledge, this is the first automatic framework for the estimation of lighting in general indoor scenarios from RGBDinput.

Irtiza Hasan, Francesco Setti, Theodore Tsesmelis et al.

Recent approaches on trajectory forecasting use tracklets to predict the future positions of pedestrians exploiting Long Short Term Memory (LSTM) architectures. This paper shows that adding vislets, that is, short sequences of head pose estimations, allows to increase significantly the trajectory forecasting performance. We then propose to use vislets in a novel framework called MX-LSTM, capturing the interplay between tracklets and vislets thanks to a joint unconstrained optimization of full covariance matrices during the LSTM backpropagation. At the same time, MX-LSTM predicts the future head poses, increasing the standard capabilities of the long-term trajectory forecasting approaches. With standard head pose estimators and an attentional-based social pooling, MX-LSTM scores the new trajectory forecasting state-of-the-art in all the considered datasets (Zara01, Zara02, UCY, and TownCentre) with a dramatic margin when the pedestrians slow down, a case where most of the forecasting approaches struggle to provide an accurate solution.