Counterfeiting affects diverse industries, including pharmaceuticals, electronics, and food, posing serious health and economic risks. Printable unclonable codes, such as Copy Detection Patterns (CDPs), are widely used as an anti-counterfeiting measure and are applied to products and packaging. However, the increasing availability of high-resolution printing and scanning devices, along with advances in generative deep learning, undermines traditional authentication systems, which often fail to distinguish high-quality counterfeits from genuine prints. In this work, we propose a diffusion-based authentication framework that jointly leverages the original binary template, the printed CDP, and a representation of printer identity that captures relevant semantic information. Formulating authentication as multi-class printer classification over printer signatures lets our model capture fine-grained, device-specific features via spatial and textual conditioning. We extend ControlNet by repurposing the denoising process for class-conditioned noise prediction, enabling effective printer classification. On the Indigo 1 × 1 Base dataset, our method outperforms traditional similarity metrics and prior deep learning approaches. Results show the framework generalizes to counterfeit types unseen during training.

In an era where climate change aggravates environmental uncertainties, the identification and detection of event precursors are becoming crucial to mitigate the impacts of disastrous natural hazards. While classical sensors such as interferometric lasers or seismometers are reliable, their widespread deployment is often hindered by logistical and economic barriers, leaving numerous blind spots. Time-lapse cameras, which already provide cost-effective, high-resolution visual context to such sensors, present a promising alternative. However, processing their output automatically faces significant challenges, notably linked to extreme shape and lighting variations. Overcoming those issues is essential to deploy them at large-scale as a monitoring tool. This paper introduces a novel sub-task of change detection, namely volumetric change detection, applied to time-lapse cameras and slope instabilities. We conduct a comprehensive review of state-of-the-art change detection methods and related tasks, analyze their core components and assess their applicability to this context. To that end, we introduce the new dataset SeracFallDet, which contains serac fall annotations and has been thoroughly annotated to meet the latter demand. Through generalization experiments, we demonstrate that dense and semi-dense feature matching, although not trained specifically for this task, exhibit robust performance. Alternatively, supervised approaches struggle with data scarcity and annotation imbalance. This suggests that hybrid methods may offer a path forward by leveraging the strengths of both tasks. These findings highlight the potential of feature matching techniques and the need for further innovation to overcome the challenges of real-world deployment in environmental monitoring.

The recent advances in deep neural networks have led us to revisit the safety of current mechanisms for the authentification of people’s identities and validating the provenance of traded goods. This paper studies the problem of identifying counterfeits of Copy Detection Patterns (CDPs). CDPs are used on product packaging to identify counterfeiting by relying on the information loss principle. Current verification techniques rely on a supervised machine learning paradigm, which needs examples from real and fake CDPs to learn a model capable of differentiating both classes of CDPs. This paper proposes an unsupervised forensic approach that is capable of training an Energy-Based Model for detecting out-of-distribution (OOD) images (fake CDP examples in our case) using only original CDP images. This paper also introduces a novel thresholding technique that only requires original CDPs for threshold value selection. We validate our approach on the Indigo dataset, with results demonstrating comparable counterfeit detection capabilities to prior work but using a method trained on less data.

Self-driving cars (SDC) are already present in certain cities of the world as part of robot taxi services. A few self-driving capabilities are also becoming more frequent in high-end consumer vehicles. But even though several datasets exist to develop methods that can enable a car to “see”, few datasets depict the actions of the occupants of an SDC in outdoor conditions. This work proposes HopInAndAction, a public dataset to evaluate action classification methods on videos of people realizing non-driving activities in the cockpit of an SDC. The dataset contains RGB recordings of 31 people carrying out 19 action classes based on 7 daily objects (e.g., telephone, newspaper, or tablet). Recordings are done in a vehicle driving autonomously in outdoor conditions and illustrate a varied set of monitoring conditions. For instance, strong and varying illumination and action occlusions due to sub-optimal sensor placement or self-body occlusion. Preliminary experiments show that HopInAndAction presents challenging monitoring conditions to the evaluated methods, as they struggle to discern between actions with similar appearances or to identify a person-object interaction in a scene with multiple objects.

Perimeter intrusion detection (PID) deals with the detection of intruders displacing in a protected perimeter. In the video surveillance domain, deep learning has shown tremendous progresses. Existing deep learning based PID systems (PIDS) are supervised and thus require a lot of annotated data. However, since intrusions are rare events, there are very few positives in datasets, thus making them highly imbalanced. Furthermore, a PIDS must adapt to varying real-life scene dynamics, like weather, light, environmental conditions, etc. To address these issues, we propose an autoencoder-based, end-to-end trainable, unsupervised PIDS with a module that can adapt to long-term variations in scene dynamics. Our results show competitive performance of the proposed system on the standard i-LIDS dataset.

In an industrial environment, object detection is a challenging task due to the absence of real images and real-time requirements for the object detector, usually embedded in a mobile device. Using 3D models, it is however possible to create a synthetic dataset to train a neural network, although the performance on real images is limited by the domain gap. In this paper, we study the performance of a Convolutional Neural Network (CNN) designed to detect objects in real-time: Single-Shot Detector (SSD) with a MobileNet backbone. We train SSD with synthetic images only, and apply extensive data augmentation to reduce the domain gap between synthetic and real images. On the T-LESS dataset, SSD performs better than Mask R-CNN trained on the same synthetic images, with MobileNet-V2 and MobileNet-V3 Large as backbone. Our results also show the huge improvement enabled by an adequate augmentation strategy.

In this work, we explore a post-processing method to enhance coarse semantic segmentation of historical aerial images. We propose to use deep edges to generate semantically meaningful superpixels that we integrate as additional pairwise potentials in a dense conditional random field. We apply our approach on very high resolution images acquired between 1975 and 1995 and annotated with land use land cover labels. Results show the interest of our approach compared to other post-processing methods.

We present an unsupervised colorization framework to improve both the visualization and the automatic land use classification of historical aerial images. We introduce a novel algorithm built upon a cyclic generative adversarial neural network and a texture replacement method to homogeneously and automatically colorize unpaired VHR images. We apply our framework on historical aerial images acquired in France between 1970 and 1990. We demonstrate that our approach helps to disentangle hard to classify land use classes and hence improves the overall land use classification.

The land cover reconstruction from monochromatic historical aerial images is a challenging task that has recently attracted an increasing interest from the scientific community with the proliferation of large-scale epidemiological studies involving retrospective analysis of spatial patterns. However, the efforts made by the computer vision community in remote-sensing applications are mostly focused on prospective approaches through the analysis of high-resolution multi-spectral data acquired by the advanced spatial programs. Hence, four contributions are proposed in this paper. They aim at providing a comparison basis for the future development of computer vision algorithms applied to the automation of the land cover reconstruction from monochromatic historical aerial images. First, a new multi-scale multi-date dataset composed of 4.9 million non-overlapping annotated patches of the France territory between 1970 and 1990 has been created with the help of geography experts. This dataset has been named HistAerial. Second, an extensive comparison study of the state-of-the-art texture features extraction and classification algorithms, including deep convolutional neural networks (DCNNs), has been performed. It is presented in the form of an evaluation. Third, a novel low-dimensional local texture filter named rotated-corner local binary pattern (R-CRLBP) is presented as a simplification of the binary gradient contours filter through the use of an orthogonal combination representation. Finally, a novel combination of low-dimensional texture descriptors, including the R-CRLBP filter, is introduced as a light combination of local binary patterns (LCoLBPs). The LCoLBP filter achieved state-of-the-art results on the HistAerial dataset while conserving a relatively low-dimensional feature vector space compared with the DCNN approaches (17 times shorter).

The recognition of complex actions is still a challenging task in Computer Vision especially in daily living scenarios, where problems like occlusion and limited field of view are very common. Recognition of Activity Daily Living (ADL) could improve the quality of life and supporting independent and healthy living of older or/and impaired people by using information and communication technologies at home, at the workplace and in public spaces. This paper proposes to embed spatio-temporal information into ontology models to improve action recognition using visual words. Actions detected by visual words are implemented as Primitive States in the scenario and then used as Components of Composite States to merge them with spatio-temporal patterns that the people display while performing ADLs. In a challenging dataset, such as SmartHome, where a high variance intra-class and low variance inter-class is present, recognition results for some actions improve in precision and recall thanks to spatial information.

Methods for action recognition have evolved considerably over the past years and can now automatically learn and recognize short term actions with satisfactory accuracy. Nonetheless, the recognition of complex activities - compositions of actions and scene objects - is still an open problem due to the complex temporal and composite structure of this category of events. Existing methods focus either on simple activities or oversimplify the modeling of complex activities by targeting only whole-part relations between its sub-parts (e.g., actions). In this paper, we propose a semi-supervised approach that learns complex activities from the temporal patterns of concept compositions (e.g., “slicing-tomato” before “pouring into-pan”). We demonstrate that our method outperforms prior work in the task of automatic modeling and recognition of complex activities learned out of the interaction of 218 distinct concepts.

Many supervised approaches report state-of-the-art results for recognizing short-term actions in manually clipped videos by utilizing fine body motion information. The main downside of these approaches is that they are not applicable in real world settings. The challenge is different when it comes to unstructured scenes and long-term videos. Unsupervised approaches have been used to model the long-term activities but the main pitfall is their limitation to handle subtle differences between similar activities since they mostly use global motion information. In this paper, we present a hybrid approach for long-term human activity recognition with more precise recognition of activities compared to unsupervised approaches. It enables processing of long-term videos by automatically clipping and performing online recognition. The performance of our approach has been tested on two Activities of Daily Living (ADL) datasets. Experimental results are promising compared to existing approaches.

Combining multimodal concept streams from heterogeneous sensors is a problem superficially explored for activity recognition. Most studies explore simple sensors in nearly perfect conditions, where temporal synchronization is guaranteed. Sophisticated fusion schemes adopt problem-specific graphical representations of events that are generally deeply linked with their training data and focused on a single sensor. This paper proposes a hybrid framework between knowledge-driven and probabilistic-driven methods for event representation and recognition. It separates semantic modeling from raw sensor data by using an intermediate semantic representation, namely concepts. It introduces an algorithm for sensor alignment that uses concept similarity as a surrogate for the inaccurate temporal information of real life scenarios. Finally, it proposes the combined use of an ontology language, to overcome the rigidity of previous approaches at model definition, and a probabilistic interpretation for ontological models, which equips the framework with a mechanism to handle noisy and ambiguous concept observations, an ability that most knowledge-driven methods lack. We evaluate our contributions in multimodal recordings of elderly people carrying out IADLs. Results demonstrated that the proposed framework outperforms baseline methods both in event recognition performance and in delimiting the temporal boundaries of event instances.

Population aging has been motivating academic research and industry to develop technologies for the improvement of older people's quality of life, medical diagnosis, and support on frailty cases. Most of available research prototypes for older people monitoring focus on fall detection or gait analysis and rely on wearable, environmental, or video sensors. We present an evaluation of a research prototype of a video monitoring system for event recognition of older people. The prototype accuracy is evaluated for the recognition of physical tasks (e.g., Up and Go test) and instrumental activities of daily living (e.g., watching TV, writing a check) of participants of a clinical protocol for Alzheimer's disease study (29 participants). The prototype uses as input a 2D RGB camera, and its performance is compared to the use of a RGB-D camera. The experimentation results show the proposed approach has a competitive performance to the use of a RGB-D camera, even outperforming it on event recognition precision. The use of a 2D-camera is advantageous, as the camera field of view can be much larger and cover an entire room where at least a couple of RGB-D cameras would be necessary.

Activity recognition has been a growing research topic in the last years and its application varies from automatic recognition of social interaction such as shaking hands, parking lot surveillance, traffic monitoring and the detection of abandoned luggage. This paper describes a probabilistic framework for uncertainty handling in a description-based event recognition approach. The proposed approach allows the flexible modeling of composite events with complex temporal constraints. It uses probability theory to provide a consistent framework for dealing with uncertain knowledge for the recognition of complex events. We validate the event recognition accuracy of the proposed algorithm on real-world videos. The experimental results show that our system can successfully recognize activities with a high recognition rate. We conclude by comparing our algorithm with the state of the art and showing how the definition of event models and the probabilistic reasoning can influence the results of real-time event recognition.