Barbara Caputo

Gabriele Berton, Riccardo Mereu, Gabriele Trivigno et al.

In this paper, we propose a new open-source benchmarking framework for Visual Geo-localization (VG) that allows to build, train, and test a wide range of commonly used architectures, with the flexibility to change individual components of a geo-localization pipeline. The purpose of this framework is twofold: i) gaining insights into how different components and design choices in a VG pipeline impact the final results, both in terms of performance (recall@N metric) and system requirements (such as execution time and memory consumption); ii) establish a systematic evaluation protocol for comparing different methods. Using the proposed framework, we perform a large suite of experiments which provide criteria for choosing backbone, aggregation and negative mining depending on the use-case and requirements. We also assess the impact of engineering techniques like pre/post-processing, data augmentation and image resizing, showing that better performance can be obtained through somewhat simple procedures: for example, downscaling the images' resolution to 80% can lead to similar results with a 36% savings in extraction time and dataset storage requirement. Code and trained models are available at https://deep-vg-bench.herokuapp.com/.

Gabriele Berton, Carlo Masone, Barbara Caputo

Visual Geo-localization (VG) is the task of estimating the position where a given photo was taken by comparing it with a large database of images of known locations. To investigate how existing techniques would perform on a real-world city-wide VG application, we build San Francisco eXtra Large, a new dataset covering a whole city and providing a wide range of challenging cases, with a size 30x bigger than the previous largest dataset for visual geo-localization. We find that current methods fail to scale to such large datasets, therefore we design a new highly scalable training technique, called CosPlace, which casts the training as a classification problem avoiding the expensive mining needed by the commonly used contrastive learning. We achieve state-of-the-art performance on a wide range of datasets and find that CosPlace is robust to heavy domain changes. Moreover, we show that, compared to the previous state-of-the-art, CosPlace requires roughly 80% less GPU memory at train time, and it achieves better results with 8x smaller descriptors, paving the way for city-wide real-world visual geo-localization. Dataset, code and trained models are available for research purposes at https://github.com/gmberton/CosPlace.

Donald Shenaj, Eros Fanì, Marco Toldo et al.

Federated Learning (FL) has recently emerged as a possible way to tackle the domain shift in real-world Semantic Segmentation (SS) without compromising the private nature of the collected data. However, most of the existing works on FL unrealistically assume labeled data in the remote clients. Here we propose a novel task (FFREEDA) in which the clients' data is unlabeled and the server accesses a source labeled dataset for pre-training only. To solve FFREEDA, we propose LADD, which leverages the knowledge of the pre-trained model by employing self-supervision with ad-hoc regularization techniques for local training and introducing a novel federated clustered aggregation scheme based on the clients' style. Our experiments show that our algorithm is able to efficiently tackle the new task outperforming existing approaches. The code is available at https://github.com/Erosinho13/LADD.

Gabriele Berton, Gabriele Trivigno, Barbara Caputo et al.

Visual Place Recognition is a task that aims to predict the place of an image (called query) based solely on its visual features. This is typically done through image retrieval, where the query is matched to the most similar images from a large database of geotagged photos, using learned global descriptors. A major challenge in this task is recognizing places seen from different viewpoints. To overcome this limitation, we propose a new method, called EigenPlaces, to train our neural network on images from different point of views, which embeds viewpoint robustness into the learned global descriptors. The underlying idea is to cluster the training data so as to explicitly present the model with different views of the same points of interest. The selection of this points of interest is done without the need for extra supervision. We then present experiments on the most comprehensive set of datasets in literature, finding that EigenPlaces is able to outperform previous state of the art on the majority of datasets, while requiring 60\% less GPU memory for training and using 50\% smaller descriptors. The code and trained models for EigenPlaces are available at {\small{\url{https://github.com/gmberton/EigenPlaces}}}, while results with any other baseline can be computed with the codebase at {\small{\url{https://github.com/gmberton/auto_VPR}}}.

Shyam Nandan Rai, Fabio Cermelli, Dario Fontanel et al.

Anomaly segmentation is a critical task for driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in high uncertainty around the objects' boundaries and numerous false positives. We propose a paradigm change by shifting from a per-pixel classification to a mask classification. Our mask-based method, Mask2Anomaly, demonstrates the feasibility of integrating an anomaly detection method in a mask-classification architecture. Mask2Anomaly includes several technical novelties that are designed to improve the detection of anomalies in masks: i) a global masked attention module to focus individually on the foreground and background regions; ii) a mask contrastive learning that maximizes the margin between an anomaly and known classes; and iii) a mask refinement solution to reduce false positives. Mask2Anomaly achieves new state-of-the-art results across a range of benchmarks, both in the per-pixel and component-level evaluations. In particular, Mask2Anomaly reduces the average false positives rate by 60% wrt the previous state-of-the-art. Github page: https://github.com/shyam671/Mask2Anomaly-Unmasking-Anomalies-in-Road-Scene-Segmentation.

Riccardo Mereu, Gabriele Trivigno, Gabriele Berton et al.

In robotics, Visual Place Recognition is a continuous process that receives as input a video stream to produce a hypothesis of the robot's current position within a map of known places. This task requires robust, scalable, and efficient techniques for real applications. This work proposes a detailed taxonomy of techniques using sequential descriptors, highlighting different mechanism to fuse the information from the individual images. This categorization is supported by a complete benchmark of experimental results that provides evidence on the strengths and weaknesses of these different architectural choices. In comparison to existing sequential descriptors methods, we further investigate the viability of Transformers instead of CNN backbones, and we propose a new ad-hoc sequence-level aggregator called SeqVLAD, which outperforms prior state of the art on different datasets. The code is available at https://github.com/vandal-vpr/vg-transformers.

Fabio Cermelli, Antonino Geraci, Dario Fontanel et al.

Despite the recent advances in the field of object detection, common architectures are still ill-suited to incrementally detect new categories over time. They are vulnerable to catastrophic forgetting: they forget what has been already learned while updating their parameters in absence of the original training data. Previous works extended standard classification methods in the object detection task, mainly adopting the knowledge distillation framework. However, we argue that object detection introduces an additional problem, which has been overlooked. While objects belonging to new classes are learned thanks to their annotations, if no supervision is provided for other objects that may still be present in the input, the model learns to associate them to background regions. We propose to handle these missing annotations by revisiting the standard knowledge distillation framework. Our approach outperforms current state-of-the-art methods in every setting of the Pascal-VOC dataset. We further propose an extension to instance segmentation, outperforming the other baselines. Code can be found here: https://github.com/fcdl94/MMA

Niccolò Cavagnero, Luca Robbiano, Barbara Caputo et al.

In the last decade, most research in Machine Learning contributed to the improvement of existing models, with the aim of increasing the performance of neural networks for the solution of a variety of different tasks. However, such advancements often come at the cost of an increase of model memory and computational requirements. This represents a significant limitation for the deployability of research output in realistic settings, where the cost, the energy consumption, and the complexity of the framework play a crucial role. To solve this issue, the designer should search for models that maximise the performance while limiting its footprint. Typical approaches to reach this goal rely either on manual procedures, which cannot guarantee the optimality of the final design, or upon Neural Architecture Search algorithms to automatise the process, at the expenses of extremely high computational time. This paper provides a solution for the fast identification of a neural network that maximises the model accuracy while preserving size and computational constraints typical of tiny devices. Our approach, named FreeREA, is a custom cell-based evolution NAS algorithm that exploits an optimised combination of training-free metrics to rank architectures during the search, thus without need of model training. Our experiments, carried out on the common benchmarks NAS-Bench-101 and NATS-Bench, demonstrate that i) FreeREA is a fast, efficient, and effective search method for models automatic design; ii) it outperforms State of the Art training-based and training-free techniques in all the datasets and benchmarks considered, and iii) it can easily generalise to constrained scenarios, representing a competitive solution for fast Neural Architecture Search in generic constrained applications. The code is available at \url{https://github.com/NiccoloCavagnero/FreeREA}.

Giovanni Barbarani, Mohamad Mostafa, Hajali Bayramov et al.

Visual Place Recognition is a task that aims to predict the coordinates of an image (called query) based solely on visual clues. Most commonly, a retrieval approach is adopted, where the query is matched to the most similar images from a large database of geotagged photos, using learned global descriptors. Despite recent advances, recognizing the same place when the query comes from a significantly different distribution is still a major hurdle for state of the art retrieval methods. Examples are heavy illumination changes (e.g. night-time images) or substantial occlusions (e.g. transient objects). In this work we explore whether re-ranking methods based on spatial verification can tackle these challenges, following the intuition that local descriptors are inherently more robust than global features to domain shifts. To this end, we provide a new, comprehensive benchmark on current state of the art models. We also introduce two new demanding datasets with night and occluded queries, to be matched against a city-wide database. Code and datasets are available at https://github.com/gbarbarani/re-ranking-for-VPR.

Gabriele Goletto, Mirco Planamente, Barbara Caputo et al.

To enable a safe and effective human-robot cooperation, it is crucial to develop models for the identification of human activities. Egocentric vision seems to be a viable solution to solve this problem, and therefore many works provide deep learning solutions to infer human actions from first person videos. However, although very promising, most of these do not consider the major challenges that comes with a realistic deployment, such as the portability of the model, the need for real-time inference, and the robustness with respect to the novel domains (i.e., new spaces, users, tasks). With this paper, we set the boundaries that egocentric vision models should consider for realistic applications, defining a novel setting of egocentric action recognition in the wild, which encourages researchers to develop novel, applications-aware solutions. We also present a new model-agnostic technique that enables the rapid repurposing of existing architectures in this new context, demonstrating the feasibility to deploy a model on a tiny device (Jetson Nano) and to perform the task directly on the edge with very low energy consumption (2.4W on average at 50 fps). The code is publicly available at: https://github.com/EgocentricVision/EgoWild.

Debora Caldarola, Barbara Caputo, Marco Ciccone

Models trained in federated settings often suffer from degraded performances and fail at generalizing, especially when facing heterogeneous scenarios. In this work, we investigate such behavior through the lens of geometry of the loss and Hessian eigenspectrum, linking the model's lack of generalization capacity to the sharpness of the solution. Motivated by prior studies connecting the sharpness of the loss surface and the generalization gap, we show that i) training clients locally with Sharpness-Aware Minimization (SAM) or its adaptive version (ASAM) and ii) averaging stochastic weights (SWA) on the server-side can substantially improve generalization in Federated Learning and help bridging the gap with centralized models. By seeking parameters in neighborhoods having uniform low loss, the model converges towards flatter minima and its generalization significantly improves in both homogeneous and heterogeneous scenarios. Empirical results demonstrate the effectiveness of those optimizers across a variety of benchmark vision datasets (e.g. CIFAR10/100, Landmarks-User-160k, IDDA) and tasks (large scale classification, semantic segmentation, domain generalization).

Chiara Plizzari, Toby Perrett, Barbara Caputo et al.

We propose and address a new generalisation problem: can a model trained for action recognition successfully classify actions when they are performed within a previously unseen scenario and in a previously unseen location? To answer this question, we introduce the Action Recognition Generalisation Over scenarios and locations dataset (ARGO1M), which contains 1.1M video clips from the large-scale Ego4D dataset, across 10 scenarios and 13 locations. We demonstrate recognition models struggle to generalise over 10 proposed test splits, each of an unseen scenario in an unseen location. We thus propose CIR, a method to represent each video as a Cross-Instance Reconstruction of videos from other domains. Reconstructions are paired with text narrations to guide the learning of a domain generalisable representation. We provide extensive analysis and ablations on ARGO1M that show CIR outperforms prior domain generalisation works on all test splits. Code and data: https://chiaraplizz.github.io/what-can-a-cook/.

Shyam Nandan Rai, Fabio Cermelli, Barbara Caputo et al.

Segmenting unknown or anomalous object instances is a critical task in autonomous driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in high uncertainty around the objects' boundaries and numerous false positives. We propose a paradigm change by shifting from a per-pixel classification to a mask classification. Our mask-based method, Mask2Anomaly, demonstrates the feasibility of integrating a mask-classification architecture to jointly address anomaly segmentation, open-set semantic segmentation, and open-set panoptic segmentation. Mask2Anomaly includes several technical novelties that are designed to improve the detection of anomalies/unknown objects: i) a global masked attention module to focus individually on the foreground and background regions; ii) a mask contrastive learning that maximizes the margin between an anomaly and known classes; iii) a mask refinement solution to reduce false positives; and iv) a novel approach to mine unknown instances based on the mask-architecture properties. By comprehensive qualitative and qualitative evaluation, we show Mask2Anomaly achieves new state-of-the-art results across the benchmarks of anomaly segmentation, open-set semantic segmentation, and open-set panoptic segmentation.

Antonio Tavera, Edoardo Arnaudo, Carlo Masone et al.

In this paper, we investigate the problem of Semantic Segmentation for agricultural aerial imagery. We observe that the existing methods used for this task are designed without considering two characteristics of the aerial data: (i) the top-down perspective implies that the model cannot rely on a fixed semantic structure of the scene, because the same scene may be experienced with different rotations of the sensor; (ii) there can be a strong imbalance in the distribution of semantic classes because the relevant objects of the scene may appear at extremely different scales (e.g., a field of crops and a small vehicle). We propose a solution to these problems based on two ideas: (i) we use together a set of suitable augmentation and a consistency loss to guide the model to learn semantic representations that are invariant to the photometric and geometric shifts typical of the top-down perspective (Augmentation Invariance); (ii) we use a sampling method (Adaptive Sampling) that selects the training images based on a measure of pixel-wise distribution of classes and actual network confidence. With an extensive set of experiments conducted on the Agriculture-Vision dataset, we demonstrate that our proposed strategies improve the performance of the current state-of-the-art method.

Edoardo Arnaudo, Antonio Tavera, Fabrizio Dominici et al.

We investigate the task of unsupervised domain adaptation in aerial semantic segmentation and discover that the current state-of-the-art algorithms designed for autonomous driving based on domain mixing do not translate well to the aerial setting. This is due to two factors: (i) a large disparity in the extension of the semantic categories, which causes a domain imbalance in the mixed image, and (ii) a weaker structural consistency in aerial scenes than in driving scenes since the same scene might be viewed from different perspectives and there is no well-defined and repeatable structure of the semantic elements in the images. Our solution to these problems is composed of: (i) a new mixing strategy for aerial segmentation across domains called Hierarchical Instance Mixing (HIMix), which extracts a set of connected components from each semantic mask and mixes them according to a semantic hierarchy and, (ii) a twin-head architecture in which two separate segmentation heads are fed with variations of the same images in a contrastive fashion to produce finer segmentation maps. We conduct extensive experiments on the LoveDA benchmark, where our solution outperforms the current state-of-the-art.

Simone Alberto Peirone, Gabriele Goletto, Mirco Planamente et al.

Human activities exhibit a strong correlation between actions and the places where these are performed, such as washing something at a sink. More specifically, in daily living environments we may identify particular locations, hereinafter named activity-centric zones, which may afford a set of homogeneous actions. Their knowledge can serve as a prior to favor vision models to recognize human activities. However, the appearance of these zones is scene-specific, limiting the transferability of this prior information to unfamiliar areas and domains. This problem is particularly relevant in egocentric vision, where the environment takes up most of the image, making it even more difficult to separate the action from the context. In this paper, we discuss the importance of decoupling the domain-specific appearance of activity-centric zones from their universal, domain-agnostic representations, and show how the latter can improve the cross-domain transferability of Egocentric Action Recognition (EAR) models. We validate our solution on the EPIC-Kitchens-100 and Argo1M datasets

Dario Fontanel, Matteo Tarantino, Fabio Cermelli et al.

Object detection methods have witnessed impressive improvements in the last years thanks to the design of novel neural network architectures and the availability of large scale datasets. However, current methods have a significant limitation: they are able to detect only the classes observed during training time, that are only a subset of all the classes that a detector may encounter in the real world. Furthermore, the presence of unknown classes is often not considered at training time, resulting in methods not even able to detect that an unknown object is present in the image. In this work, we address the problem of detecting unknown objects, known as open-set object detection. We propose a novel training strategy, called UNKAD, able to predict unknown objects without requiring any annotation of them, exploiting non annotated objects that are already present in the background of training images. In particular, exploiting the four-steps training strategy of Faster R-CNN, UNKAD first identifies and pseudo-labels unknown objects and then uses the pseudo-annotations to train an additional unknown class. While UNKAD can directly detect unknown objects, we further combine it with previous unknown detection techniques, showing that it improves their performance at no costs.

Eros Fanì, Marco Ciccone, Barbara Caputo

We propose FedDrive v2, an extension of the Federated Learning benchmark for Semantic Segmentation in Autonomous Driving. While the first version aims at studying the effect of domain shift of the visual features across clients, in this work, we focus on the distribution skewness of the labels. We propose six new federated scenarios to investigate how label skewness affects the performance of segmentation models and compare it with the effect of domain shift. Finally, we study the impact of using the domain information during testing. Official website: https://feddrive.github.io

Debora Caldarola, Barbara Caputo, Marco Ciccone

Federated Learning (FL) aims to learn a global model from distributed users while protecting their privacy. However, when data are distributed heterogeneously the learning process becomes noisy, unstable, and biased towards the last seen clients' data, slowing down convergence. To address these issues and improve the robustness and generalization capabilities of the global model, we propose WIMA (Window-based Model Averaging). WIMA aggregates global models from different rounds using a window-based approach, effectively capturing knowledge from multiple users and reducing the bias from the last ones. By adopting a windowed view on the rounds, WIMA can be applied from the initial stages of training. Importantly, our method introduces no additional communication or client-side computation overhead. Our experiments demonstrate the robustness of WIMA against distribution shifts and bad client sampling, resulting in smoother and more stable learning trends. Additionally, WIMA can be easily integrated with state-of-the-art algorithms. We extensively evaluate our approach on standard FL benchmarks, demonstrating its effectiveness.

Amirshayan Nasirimajd, Simone Alberto Peirone, Chiara Plizzari et al.

This report presents the technical details of our approach for the EPIC-Kitchens-100 Unsupervised Domain Adaptation (UDA) Challenge in Action Recognition. Our approach is based on the idea that the order in which actions are performed is similar between the source and target domains. Based on this, we generate a modified sequence by randomly combining actions from the source and target domains. As only unlabelled target data are available under the UDA setting, we use a standard pseudo-labeling strategy for extracting action labels for the target. We then ask the network to predict the resulting action sequence. This allows to integrate information from both domains during training and to achieve better transfer results on target. Additionally, to better incorporate sequence information, we use a language model to filter unlikely sequences. Lastly, we employed a co-occurrence matrix to eliminate unseen combinations of verbs and nouns. Our submission, labeled as 'sshayan', can be found on the leaderboard, where it currently holds the 2nd position for 'verb' and the 4th position for both 'noun' and 'action'.

Gabriele Trivigno, Gabriele Berton, Juan Aragon et al.

Visual Place recognition is commonly addressed as an image retrieval problem. However, retrieval methods are impractical to scale to large datasets, densely sampled from city-wide maps, since their dimension impact negatively on the inference time. Using approximate nearest neighbour search for retrieval helps to mitigate this issue, at the cost of a performance drop. In this paper we investigate whether we can effectively approach this task as a classification problem, thus bypassing the need for a similarity search. We find that existing classification methods for coarse, planet-wide localization are not suitable for the fine-grained and city-wide setting. This is largely due to how the dataset is split into classes, because these methods are designed to handle a sparse distribution of photos and as such do not consider the visual aliasing problem across neighbouring classes that naturally arises in dense scenarios. Thus, we propose a partitioning scheme that enables a fast and accurate inference, preserving a simple learning procedure, and a novel inference pipeline based on an ensemble of novel classifiers that uses the prototypes learned via an angular margin loss. Our method, Divide&Classify (D&C), enjoys the fast inference of classification solutions and an accuracy competitive with retrieval methods on the fine-grained, city-wide setting. Moreover, we show that D&C can be paired with existing retrieval pipelines to speed up computations by over 20 times while increasing their recall, leading to new state-of-the-art results.

Gabriele Trivigno, Carlo Masone, Barbara Caputo et al.

Pose refinement is an interesting and practically relevant research direction. Pose refinement can be used to (1) obtain a more accurate pose estimate from an initial prior (e.g., from retrieval), (2) as pre-processing, i.e., to provide a better starting point to a more expensive pose estimator, (3) as post-processing of a more accurate localizer. Existing approaches focus on learning features / scene representations for the pose refinement task. This involves training an implicit scene representation or learning features while optimizing a camera pose-based loss. A natural question is whether training specific features / representations is truly necessary or whether similar results can be already achieved with more generic features. In this work, we present a simple approach that combines pre-trained features with a particle filter and a renderable representation of the scene. Despite its simplicity, it achieves state-of-the-art results, demonstrating that one can easily build a pose refiner without the need for specific training. The code is at https://github.com/ga1i13o/mcloc_poseref

Gabriele Rosi, Fabio Cermelli, Carlo Masone et al.

Segmenting images is critical for visual understanding but demands extensive pixel-level annotations. Foundational models have enabled new paradigms for predicting new classes guided by textual prompts, without annotations from the target domain. Yet, on specialized target domains, far from the original pre-training, their performance degrades. We study the errors of existing methods under such domain-shift, finding that misclassification rather than mask generation is the main culprit. To address this, we introduce the novel problem of Few-Shot Visual Adaptation for text-prompted Segmentation. This kind of adaptation has been largely studied for image classification, but it remains unexplored for segmentation. We tackle this task with Prototype Adaptation (PrAda), a novel, parameter-efficient method that adapts a frozen text-prompted segmentation model. Our approach learns class-specific prototypes by combining fine-grained pixel features and high-level transformer representations, which are then fused with the original text-based predictions through a learned importance factor. This preserves the model's zero-shot potential while enabling strong adaptation to new domains. Experiments across semantic, instance, and panoptic segmentation on five benchmarks demonstrate that PrAda yields significant improvements over state-of-the-art and proposed baselines.

Niccolò Cavagnero, Luca Robbiano, Francesca Pistilli et al.

Neural Networks design is a complex and often daunting task, particularly for resource-constrained scenarios typical of mobile-sized models. Neural Architecture Search is a promising approach to automate this process, but existing competitive methods require large training time and computational resources to generate accurate models. To overcome these limits, this paper contributes with: i) a novel training-free metric, named Entropic Score, to estimate model expressivity through the aggregated element-wise entropy of its activations; ii) a cyclic search algorithm to separately yet synergistically search model size and topology. Entropic Score shows remarkable ability in searching for the topology of the network, and a proper combination with LogSynflow, to search for model size, yields superior capability to completely design high-performance Hybrid Transformers for edge applications in less than 1 GPU hour, resulting in the fastest and most accurate NAS method for ImageNet classification.

Gabriele Berton, Gabriele Trivigno, Barbara Caputo et al.

Visual Place Recognition aims at recognizing previously visited places by relying on visual clues, and it is used in robotics applications for SLAM and localization. Since typically a mobile robot has access to a continuous stream of frames, this task is naturally cast as a sequence-to-sequence localization problem. Nevertheless, obtaining sequences of labelled data is much more expensive than collecting isolated images, which can be done in an automated way with little supervision. As a mitigation to this problem, we propose a novel Joint Image and Sequence Training protocol (JIST) that leverages large uncurated sets of images through a multi-task learning framework. With JIST we also introduce SeqGeM, an aggregation layer that revisits the popular GeM pooling to produce a single robust and compact embedding from a sequence of single-frame embeddings. We show that our model is able to outperform previous state of the art while being faster, using 8 times smaller descriptors, having a lighter architecture and allowing to process sequences of various lengths. Code is available at https://github.com/ga1i13o/JIST

Gabriele Berton, Alex Stoken, Barbara Caputo et al.

Astronaut photography, spanning six decades of human spaceflight, presents a unique Earth observations dataset with immense value for both scientific research and disaster response. Despite its significance, accurately localizing the geographical extent of these images, crucial for effective utilization, poses substantial challenges. Current manual localization efforts are time-consuming, motivating the need for automated solutions. We propose a novel approach - leveraging image retrieval - to address this challenge efficiently. We introduce innovative training techniques, including Year-Wise Data Augmentation and a Neutral-Aware Multi-Similarity Loss, which contribute to the development of a high-performance model, EarthLoc. We develop six evaluation datasets and perform a comprehensive benchmark comparing EarthLoc to existing methods, showcasing its superior efficiency and accuracy. Our approach marks a significant advancement in automating the localization of astronaut photography, which will help bridge a critical gap in Earth observations data. Code and datasets are available at https://github.com/gmberton/EarthLoc

Mirco Planamente, Gabriele Goletto, Gabriele Trivigno et al.

In this report, we describe the technical details of our submission to the EPIC-Kitchens-100 Unsupervised Domain Adaptation (UDA) Challenge in Action Recognition. To tackle the domain-shift which exists under the UDA setting, we first exploited a recent Domain Generalization (DG) technique, called Relative Norm Alignment (RNA). Secondly, we extended this approach to work on unlabelled target data, enabling a simpler adaptation of the model to the target distribution in an unsupervised fashion. To this purpose, we included in our framework UDA algorithms, such as multi-level adversarial alignment and attentive entropy. By analyzing the challenge setting, we notice the presence of a secondary concurrence shift in the data, which is usually called environmental bias. It is caused by the existence of different environments, i.e., kitchens. To deal with these two shifts (environmental and temporal), we extended our system to perform Multi-Source Multi-Target Domain Adaptation. Finally, we employed distinct models in our final proposal to leverage the potential of popular video architectures, and we introduced two more losses for the ensemble adaptation. Our submission (entry 'plnet') is visible on the leaderboard and ranked in 2nd position for 'verb', and in 3rd position for both 'noun' and 'action'.

Edoardo Arnaudo, Fabio Cermelli, Antonio Tavera et al.

Incremental learning represents a crucial task in aerial image processing, especially given the limited availability of large-scale annotated datasets. A major issue concerning current deep neural architectures is known as catastrophic forgetting, namely the inability to faithfully maintain past knowledge once a new set of data is provided for retraining. Over the years, several techniques have been proposed to mitigate this problem for image classification and object detection. However, only recently the focus has shifted towards more complex downstream tasks such as instance or semantic segmentation. Starting from incremental-class learning for semantic segmentation tasks, our goal is to adapt this strategy to the aerial domain, exploiting a peculiar feature that differentiates it from natural images, namely the orientation. In addition to the standard knowledge distillation approach, we propose a contrastive regularization, where any given input is compared with its augmented version (i.e. flipping and rotations) in order to minimize the difference between the segmentation features produced by both inputs. We show the effectiveness of our solution on the Potsdam dataset, outperforming the incremental baseline in every test. Code available at: https://github.com/edornd/contrastive-distillation.

Fabio Cermelli, Dario Fontanel, Antonio Tavera et al.

Although existing semantic segmentation approaches achieve impressive results, they still struggle to update their models incrementally as new categories are uncovered. Furthermore, pixel-by-pixel annotations are expensive and time-consuming. This paper proposes a novel framework for Weakly Incremental Learning for Semantic Segmentation, that aims at learning to segment new classes from cheap and largely available image-level labels. As opposed to existing approaches, that need to generate pseudo-labels offline, we use an auxiliary classifier, trained with image-level labels and regularized by the segmentation model, to obtain pseudo-supervision online and update the model incrementally. We cope with the inherent noise in the process by using soft-labels generated by the auxiliary classifier. We demonstrate the effectiveness of our approach on the Pascal VOC and COCO datasets, outperforming offline weakly-supervised methods and obtaining results comparable with incremental learning methods with full supervision. Code can be found at https://github.com/fcdl94/WILSON.

Antonio Tavera, Carlo Masone, Barbara Caputo

Semantic segmentation models have reached remarkable performance across various tasks. However, this performance is achieved with extremely large models, using powerful computational resources and without considering training and inference time. Real-world applications, on the other hand, necessitate models with minimal memory demands, efficient inference speed, and executable with low-resources embedded devices, such as self-driving vehicles. In this paper, we look at the challenge of real-time semantic segmentation across domains, and we train a model to act appropriately on real-world data even though it was trained on a synthetic realm. We employ a new lightweight and shallow discriminator that was specifically created for this purpose. To the best of our knowledge, we are the first to present a real-time adversarial approach for assessing the domain adaption problem in semantic segmentation. We tested our framework in the two standard protocol: GTA5 to Cityscapes and SYNTHIA to Cityscapes. Code is available at: https://github.com/taveraantonio/RTDA.

Gabriele Berton, Carlo Masone, Valerio Paolicelli et al.

In this paper we propose a novel method for image matching based on dense local features and tailored for visual geolocalization. Dense local features matching is robust against changes in illumination and occlusions, but not against viewpoint shifts which are a fundamental aspect of geolocalization. Our method, called GeoWarp, directly embeds invariance to viewpoint shifts in the process of extracting dense features. This is achieved via a trainable module which learns from the data an invariance that is meaningful for the task of recognizing places. We also devise a new self-supervised loss and two new weakly supervised losses to train this module using only unlabeled data and weak labels. GeoWarp is implemented efficiently as a re-ranking method that can be easily embedded into pre-existing visual geolocalization pipelines. Experimental validation on standard geolocalization benchmarks demonstrates that GeoWarp boosts the accuracy of state-of-the-art retrieval architectures. The code and trained models are available at https://github.com/gmberton/geo_warp

Dario Fontanel, Fabio Cermelli, Massimiliano Mancini et al.

Robotic visual systems operating in the wild must act in unconstrained scenarios, under different environmental conditions while facing a variety of semantic concepts, including unknown ones. To this end, recent works tried to empower visual object recognition methods with the capability to i) detect unseen concepts and ii) extended their knowledge over time, as images of new semantic classes arrive. This setting, called Open World Recognition (OWR), has the goal to produce systems capable of breaking the semantic limits present in the initial training set. However, this training set imposes to the system not only its own semantic limits, but also environmental ones, due to its bias toward certain acquisition conditions that do not necessarily reflect the high variability of the real-world. This discrepancy between training and test distribution is called domain-shift. This work investigates whether OWR algorithms are effective under domain-shift, presenting the first benchmark setup for assessing fairly the performances of OWR algorithms, with and without domain-shift. We then use this benchmark to conduct analyses in various scenarios, showing how existing OWR algorithms indeed suffer a severe performance degradation when train and test distributions differ. Our analysis shows that this degradation is only slightly mitigated by coupling OWR with domain generalization techniques, indicating that the mere plug-and-play of existing algorithms is not enough to recognize new and unknown categories in unseen domains. Our results clearly point toward open issues and future research directions, that need to be investigated for building robot visual systems able to function reliably under these challenging yet very real conditions. Code available at https://github.com/DarioFontanel/OWR-VisualDomains

Silvia Bucci, Francesco Cappio Borlino, Barbara Caputo et al.

Vision systems trained in closed-world scenarios fail when presented with new environmental conditions, new data distributions, and novel classes at deployment time. How to move towards open-world learning is a long-standing research question. The existing solutions mainly focus on specific aspects of the problem (single domain Open-Set, multi-domain Closed-Set), or propose complex strategies which combine several losses and manually tuned hyperparameters. In this work, we tackle multi-source Open-Set domain adaptation by introducing HyMOS: a straightforward model that exploits the power of contrastive learning and the properties of its hyperspherical feature space to correctly predict known labels on the target, while rejecting samples belonging to any unknown class. HyMOS includes style transfer among the instance transformations of contrastive learning to get domain invariance while avoiding the risk of negative-transfer. A self-paced threshold is defined on the basis of the observed data distribution and updates online during training, allowing to handle the known-unknown separation. We validate our method over three challenging datasets. The obtained results show that HyMOS outperforms several competitors, defining the new state-of-the-art. Our code is available at https://github.com/silvia1993/HyMOS.

Dario Fontanel, Fabio Cermelli, Massimiliano Mancini et al.

Traditional semantic segmentation methods can recognize at test time only the classes that are present in the training set. This is a significant limitation, especially for semantic segmentation algorithms mounted on intelligent autonomous systems, deployed in realistic settings. Regardless of how many classes the system has seen at training time, it is inevitable that unexpected, unknown objects will appear at test time. The failure in identifying such anomalies may lead to incorrect, even dangerous behaviors of the autonomous agent equipped with such segmentation model when deployed in the real world. Current state of the art of anomaly segmentation uses generative models, exploiting their incapability to reconstruct patterns unseen during training. However, training these models is expensive, and their generated artifacts may create false anomalies. In this paper we take a different route and we propose to address anomaly segmentation through prototype learning. Our intuition is that anomalous pixels are those that are dissimilar to all class prototypes known by the model. We extract class prototypes from the training data in a lightweight manner using a cosine similarity-based classifier. Experiments on StreetHazards show that our approach achieves the new state of the art, with a significant margin over previous works, despite the reduced computational overhead. Code is available at https://github.com/DarioFontanel/PAnS.

Matteo A. Senese, Alberto Benincasa, Barbara Caputo et al.

This paper presents our work for the ninth edition of the Dialogue System Technology Challenge (DSTC9). Our solution addresses the track number four: Simulated Interactive MultiModal Conversations. The task consists in providing an algorithm able to simulate a shopping assistant that supports the user with his/her requests. We address the task of response retrieval, that is the task of retrieving the most appropriate agent response from a pool of response candidates. Our approach makes use of a neural architecture based on transformer with a multi-attentive structure that conditions the response of the agent on the request made by the user and on the product the user is referring to. Final experiments on the SIMMC Fashion Dataset show that our approach achieves the second best scores on all the retrieval metrics defined by the organizers. The source code is available at https://github.com/D2KLab/dstc9-SIMMC.

Gabriele Moreno Berton, Valerio Paolicelli, Carlo Masone et al.

We address the task of cross-domain visual place recognition, where the goal is to geolocalize a given query image against a labeled gallery, in the case where the query and the gallery belong to different visual domains. To achieve this, we focus on building a domain robust deep network by leveraging over an attention mechanism combined with few-shot unsupervised domain adaptation techniques, where we use a small number of unlabeled target domain images to learn about the target distribution. With our method, we are able to outperform the current state of the art while using two orders of magnitude less target domain images. Finally we propose a new large-scale dataset for cross-domain visual place recognition, called SVOX. The pytorch code is available at https://github.com/valeriopaolicelli/AdAGeo .

Gabriele Berton, Gabriele Goletto, Gabriele Trivigno et al.

Precise, pixel-wise geolocalization of astronaut photography is critical to unlocking the potential of this unique type of remotely sensed Earth data, particularly for its use in disaster management and climate change research. Recent works have established the Astronaut Photography Localization task, but have either proved too costly for mass deployment or generated too coarse a localization. Thus, we present EarthMatch, an iterative homography estimation method that produces fine-grained localization of astronaut photographs while maintaining an emphasis on speed. We refocus the astronaut photography benchmark, AIMS, on the geolocalization task itself, and prove our method's efficacy on this dataset. In addition, we offer a new, fair method for image matcher comparison, and an extensive evaluation of different matching models within our localization pipeline. Our method will enable fast and accurate localization of the 4.5 million and growing collection of astronaut photography of Earth. Webpage with code and data at https://earthloc-and-earthmatch.github.io

Debora Caldarola, Pietro Cagnasso, Barbara Caputo et al.

Federated learning (FL) enables collaborative model training with privacy preservation. Data heterogeneity across edge devices (clients) can cause models to converge to sharp minima, negatively impacting generalization and robustness. Recent approaches use client-side sharpness-aware minimization (SAM) to encourage flatter minima, but the discrepancy between local and global loss landscapes often undermines their effectiveness, as optimizing for local sharpness does not ensure global flatness. This work introduces FedGloSS (Federated Global Server-side Sharpness), a novel FL approach that prioritizes the optimization of global sharpness on the server, using SAM. To reduce communication overhead, FedGloSS cleverly approximates sharpness using the previous global gradient, eliminating the need for additional client communication. Our extensive evaluations demonstrate that FedGloSS consistently reaches flatter minima and better performance compared to state-of-the-art FL methods across various federated vision benchmarks.

Claudia Cuttano, Antonio Tavera, Fabio Cermelli et al.

Many practical applications require training of semantic segmentation models on unlabelled datasets and their execution on low-resource hardware. Distillation from a trained source model may represent a solution for the first but does not account for the different distribution of the training data. Unsupervised domain adaptation (UDA) techniques claim to solve the domain shift, but in most cases assume the availability of the source data or an accessible white-box source model, which in practical applications are often unavailable for commercial and/or safety reasons. In this paper, we investigate a more challenging setting in which a lightweight model has to be trained on a target unlabelled dataset for semantic segmentation, under the assumption that we have access only to black-box source model predictions. Our method, named CoRTe, consists of (i) a pseudo-labelling function that extracts reliable knowledge from the black-box source model using its relative confidence, (ii) a pseudo label refinement method to retain and enhance the novel information learned by the student model on the target data, and (iii) a consistent training of the model using the extracted pseudo labels. We benchmark CoRTe on two synthetic-to-real settings, demonstrating remarkable results when using black-box models to transfer knowledge on lightweight models for a target data distribution.

Alessandro Licciardi, Davide Leo, Eros Fanì et al.

Federated Learning (FL) emerged as a decentralized paradigm to train models while preserving privacy. However, conventional FL struggles with data heterogeneity and class imbalance, which degrade model performance. Clustered FL balances personalization and decentralized training by grouping clients with analogous data distributions, enabling improved accuracy while adhering to privacy constraints. This approach effectively mitigates the adverse impact of heterogeneity in FL. In this work, we propose a novel clustered FL method, FedGWC (Federated Gaussian Weighting Clustering), which groups clients based on their data distribution, allowing training of a more robust and personalized model on the identified clusters. FedGWC identifies homogeneous clusters by transforming individual empirical losses to model client interactions with a Gaussian reward mechanism. Additionally, we introduce the Wasserstein Adjusted Score, a new clustering metric for FL to evaluate cluster cohesion with respect to the individual class distribution. Our experiments on benchmark datasets show that FedGWC outperforms existing FL algorithms in cluster quality and classification accuracy, validating the efficacy of our approach.

Shyam Nandan Rai, Shyamgopal Karthik, Mariana-Iuliana Georgescu et al.

With the growing deployment of autonomous driving agents, the detection and segmentation of road obstacles have become critical to ensure safe autonomous navigation. However, existing road-obstacle segmentation methods are applied on individual frames, overlooking the temporal nature of the problem, leading to inconsistent prediction maps between consecutive frames. In this work, we demonstrate that the road-obstacle segmentation task is inherently temporal, since the segmentation maps for consecutive frames are strongly correlated. To address this, we curate and adapt four evaluation benchmarks for road-obstacle video segmentation and evaluate 11 state-of-the-art image- and video-based segmentation methods on these benchmarks. Moreover, we introduce two strong baseline methods based on vision foundation models. Our approach establishes a new state-of-the-art in road-obstacle video segmentation for long-range video sequences, providing valuable insights and direction for future research.

Amirshayan Nasirimajd, Chiara Plizzari, Simone Alberto Peirone et al.

Recognizing human activities from visual inputs, particularly through a first-person viewpoint, is essential for enabling robots to replicate human behavior. Egocentric vision, characterized by cameras worn by observers, captures diverse changes in illumination, viewpoint, and environment. This variability leads to a notable drop in the performance of Egocentric Action Recognition models when tested in environments not seen during training. In this paper, we tackle these challenges by proposing a domain generalization approach for Egocentric Action Recognition. Our insight is that action sequences often reflect consistent user intent across visual domains. By leveraging action sequences, we aim to enhance the model's generalization ability across unseen environments. Our proposed method, named SeqDG, introduces a visual-text sequence reconstruction objective (SeqRec) that uses contextual cues from both text and visual inputs to reconstruct the central action of the sequence. Additionally, we enhance the model's robustness by training it on mixed sequences of actions from different domains (SeqMix). We validate SeqDG on the EGTEA and EPIC-KITCHENS-100 datasets. Results on EPIC-KITCHENS-100, show that SeqDG leads to +2.4% relative average improvement in cross-domain action recognition in unseen environments, and on EGTEA the model achieved +0.6% Top-1 accuracy over SOTA in intra-domain action recognition.

Gergely D. Németh, Eros Fanì, Yeat Jeng Ng et al.

Federated Learning (FL) enables decentralized training of machine learning models on distributed data while preserving privacy. However, in real-world FL settings, client data is often non-identically distributed and imbalanced, resulting in statistical data heterogeneity which impacts the generalization capabilities of the server's model across clients, slows convergence and reduces performance. In this paper, we address this challenge by proposing first a characterization of statistical data heterogeneity by means of 6 metrics of global and client attribute imbalance, class imbalance, and spurious correlations. Next, we create and share 7 computer vision datasets for binary and multiclass image classification tasks in Federated Learning that cover a broad range of statistical data heterogeneity and hence simulate real-world situations. Finally, we propose FEDDIVERSE, a novel client selection algorithm in FL which is designed to manage and leverage data heterogeneity across clients by promoting collaboration between clients with complementary data distributions. Experiments on the seven proposed FL datasets demonstrate FEDDIVERSE's effectiveness in enhancing the performance and robustness of a variety of FL methods while having low communication and computational overhead.

Gabriele Berton, Lorenz Junglas, Riccardo Zaccone et al.

Mesh-based scene representation offers a promising direction for simplifying large-scale hierarchical visual localization pipelines, combining a visual place recognition step based on global features (retrieval) and a visual localization step based on local features. While existing work demonstrates the viability of meshes for visual localization, the impact of using synthetic databases rendered from them in visual place recognition remains largely unexplored. In this work we investigate using dense 3D textured meshes for large-scale Visual Place Recognition (VPR). We identify a significant performance drop when using synthetic mesh-based image databases compared to real-world images for retrieval. To address this, we propose MeshVPR, a novel VPR pipeline that utilizes a lightweight features alignment framework to bridge the gap between real-world and synthetic domains. MeshVPR leverages pre-trained VPR models and is efficient and scalable for city-wide deployments. We introduce novel datasets with freely available 3D meshes and manually collected queries from Berlin, Paris, and Melbourne. Extensive evaluations demonstrate that MeshVPR achieves competitive performance with standard VPR pipelines, paving the way for mesh-based localization systems. Data, code, and interactive visualizations are available at https://meshvpr.github.io/

Eros Fanì, Raffaello Camoriano, Barbara Caputo et al.

Federated Learning (FL) methods often struggle in highly statistically heterogeneous settings. Indeed, non-IID data distributions cause client drift and biased local solutions, particularly pronounced in the final classification layer, negatively impacting convergence speed and accuracy. To address this issue, we introduce Federated Recursive Ridge Regression (Fed3R). Our method fits a Ridge Regression classifier computed in closed form leveraging pre-trained features. Fed3R is immune to statistical heterogeneity and is invariant to the sampling order of the clients. Therefore, it proves particularly effective in cross-device scenarios. Furthermore, it is fast and efficient in terms of communication and computation costs, requiring up to two orders of magnitude fewer resources than the competitors. Finally, we propose to leverage the Fed3R parameters as an initialization for a softmax classifier and subsequently fine-tune the model using any FL algorithm (Fed3R with Fine-Tuning, Fed3R+FT). Our findings also indicate that maintaining a fixed classifier aids in stabilizing the training and learning more discriminative features in cross-device settings. Official website: https://fed-3r.github.io/.

Lidia Fantauzzo, Eros Fanì, Debora Caldarola et al.

Semantic Segmentation is essential to make self-driving vehicles autonomous, enabling them to understand their surroundings by assigning individual pixels to known categories. However, it operates on sensible data collected from the users' cars; thus, protecting the clients' privacy becomes a primary concern. For similar reasons, Federated Learning has been recently introduced as a new machine learning paradigm aiming to learn a global model while preserving privacy and leveraging data on millions of remote devices. Despite several efforts on this topic, no work has explicitly addressed the challenges of federated learning in semantic segmentation for driving so far. To fill this gap, we propose FedDrive, a new benchmark consisting of three settings and two datasets, incorporating the real-world challenges of statistical heterogeneity and domain generalization. We benchmark state-of-the-art algorithms from the federated learning literature through an in-depth analysis, combining them with style transfer methods to improve their generalization ability. We demonstrate that correctly handling normalization statistics is crucial to deal with the aforementioned challenges. Furthermore, style transfer improves performance when dealing with significant appearance shifts. Official website: https://feddrive.github.io.

Fabio Cermelli, Massimiliano Mancini, Samuel Rota Buló et al.

Deep neural networks have enabled major progresses in semantic segmentation. However, even the most advanced neural architectures suffer from important limitations. First, they are vulnerable to catastrophic forgetting, i.e. they perform poorly when they are required to incrementally update their model as new classes are available. Second, they rely on large amount of pixel-level annotations to produce accurate segmentation maps. To tackle these issues, we introduce a novel incremental class learning approach for semantic segmentation taking into account a peculiar aspect of this task: since each training step provides annotation only for a subset of all possible classes, pixels of the background class exhibit a semantic shift. Therefore, we revisit the traditional distillation paradigm by designing novel loss terms which explicitly account for the background shift. Additionally, we introduce a novel strategy to initialize classifier's parameters at each step in order to prevent biased predictions toward the background class. Finally, we demonstrate that our approach can be extended to point- and scribble-based weakly supervised segmentation, modeling the partial annotations to create priors for unlabeled pixels. We demonstrate the effectiveness of our approach with an extensive evaluation on the Pascal-VOC, ADE20K, and Cityscapes datasets, significantly outperforming state-of-the-art methods.

Riccardo Zaccone, Andrea Rizzardi, Debora Caldarola et al.

Federated Learning (FL) allows training machine learning models in privacy-constrained scenarios by enabling the cooperation of edge devices without requiring local data sharing. This approach raises several challenges due to the different statistical distribution of the local datasets and the clients' computational heterogeneity. In particular, the presence of highly non-i.i.d. data severely impairs both the performance of the trained neural network and its convergence rate, increasing the number of communication rounds requested to reach a performance comparable to that of the centralized scenario. As a solution, we propose FedSeq, a novel framework leveraging the sequential training of subgroups of heterogeneous clients, i.e. superclients, to emulate the centralized paradigm in a privacy-compliant way. Given a fixed budget of communication rounds, we show that FedSeq outperforms or match several state-of-the-art federated algorithms in terms of final performance and speed of convergence. Finally, our method can be easily integrated with other approaches available in the literature. Empirical results show that combining existing algorithms with FedSeq further improves its final performance and convergence speed. We test our method on CIFAR-10 and CIFAR-100 and prove its effectiveness in both i.i.d. and non-i.i.d. scenarios.

Valerio Paolicelli, Antonio Tavera, Carlo Masone et al.

In this paper we address the task of visual place recognition (VPR), where the goal is to retrieve the correct GPS coordinates of a given query image against a huge geotagged gallery. While recent works have shown that building descriptors incorporating semantic and appearance information is beneficial, current state-of-the-art methods opt for a top down definition of the significant semantic content. Here we present the first VPR algorithm that learns robust global embeddings from both visual appearance and semantic content of the data, with the segmentation process being dynamically guided by the recognition of places through a multi-scale attention module. Experiments on various scenarios validate this new approach and demonstrate its performance against state-of-the-art methods. Finally, we propose the first synthetic-world dataset suited for both place recognition and segmentation tasks.

Chiara Plizzari, Mirco Planamente, Gabriele Goletto et al.

Event cameras are novel bio-inspired sensors, which asynchronously capture pixel-level intensity changes in the form of "events". Due to their sensing mechanism, event cameras have little to no motion blur, a very high temporal resolution and require significantly less power and memory than traditional frame-based cameras. These characteristics make them a perfect fit to several real-world applications such as egocentric action recognition on wearable devices, where fast camera motion and limited power challenge traditional vision sensors. However, the ever-growing field of event-based vision has, to date, overlooked the potential of event cameras in such applications. In this paper, we show that event data is a very valuable modality for egocentric action recognition. To do so, we introduce N-EPIC-Kitchens, the first event-based camera extension of the large-scale EPIC-Kitchens dataset. In this context, we propose two strategies: (i) directly processing event-camera data with traditional video-processing architectures (E$^2$(GO)) and (ii) using event-data to distill optical flow information (E$^2$(GO)MO). On our proposed benchmark, we show that event data provides a comparable performance to RGB and optical flow, yet without any additional flow computation at deploy time, and an improved performance of up to 4% with respect to RGB only information.