In recent years, the demand for large, diverse, and well-annotated firearm datasets has grown, particularly for advancing firearm recognition systems. However, existing datasets often lack sufficient variety in viewpoints, configurations, and model versions. To address these challenges, this paper introduces an extented version of the Ceasefire Hierarchical Firearm Dataset (eCHFD), a large image dataset of firearms consisting of over 93,000 images in 505 classes. It was constructed from more than 240 videos filmed at the Toulouse Forensics Laboratory (France) and further enriched with images from the existing CHFD dataset and additional downloaded images. Automated post-processing techniques were employed to efficiently build this large-scale dataset. In addition, a hierarchical annotation scheme validated by forensics experts was used to classify firearms at multiple levels, enabling its utility for fine-grained classification tasks. Experimental results was conducted using models such as HERBS and FGVC-PIM to demonstrate the effectiveness of the dataset.

Modern technologies have enabled illicit firearms trafficking to partially merge with cybercrime, while also allowing its off-line aspects to become increasingly complex. The online trade of firearms, their components, 3D blueprints and illicit substances carried out by criminals on both the surface Web and dark Web is increasingly difficult to address as a consequence of the exponential growth in the amount of information disseminated on the Internet. On the other hand, law enforcement agencies are confronted with significant challenges that require the development of sophisticated technological solutions capable of processing large volumes of data, identifying relevant information in a timely manner and creating networks of connections between potential criminal groups. This article presents a real-world practical system, namely the CEASEFIRE one, powered by advanced artificial intelligence technologies that can assist law enforcement personnel in addressing the above described challenges.

Given the huge level of firearms trafficking in Europe, governments, and in particular interior ministries, are actively engaged in developing artificial intelligence tools to identify firearms more effectively. Indeed, confronted with the huge number of different possible firearms, when a weapon is seized, police officers are not always able to identify it and in particular its illegal nature or the dangerousness of owning it. Consequently, we aim to develop a recognition approach based on a neural network. To this end, a training database is required. As firearms can be hierarchically classified, we have decided to build a hierarchical database. This article presents the Ceasefire Hierarchical Weapon Dataset, an RGB image dataset of firearms tailored for fine-grained image classification. It contains 260 classes ranging from 25 to hundreds of images per class, with a total of 40,789 images. In addition, a 4-level hierarchy (family, group, type, model) is provided and validated by forensic experts. The dataset was built from more than 23,000 images acquired at the Toulouse forensics laboratory (France) using frames extracted from videos filmed on site, and completed with the existing Firearm 14k dataset enriched with 2,526 downloaded images from the Web. Results of hierarchical classification using the proposed dataset are also reported.

This paper proposes blind mesh quality measures for a reconstructed 3D model. The assessment of 3D model quality is a fundamental step in the process of 3D reconstruction, to efficiently and iteratively improve the model quality. We first prove that metrics based on a reference model (extrinsic metrics) that have been shown to be correlated to subjective interpretation, are able to capture flaws that may occur while reconstructing. However, no reference is available during an iterative reconstruction process, so we study intrinsic measures on a 3D model to evaluate the quality of the 3D model being reconstructed. As expected, our results show that these intrinsic metrics give high responses in regions locally corrupted by noises.

The determination of precise skin lesion boundaries in dermoscopic images using automated methods faces many challenges, most importantly, the presence of hair, inconspicuous lesion edges and low contrast in dermoscopic images, and variability in the color, texture and shapes of skin lesions. Existing deep learning-based skin lesion segmentation algorithms are expensive in terms of computational time and memory. Consequently, running such segmentation algorithms requires a powerful GPU and high bandwidth memory, which are not available in dermoscopy devices. Thus, this article aims to achieve precise skin lesion segmentation with minimum resources: a lightweight, efficient generative adversarial network (GAN) model called SLSNet, which combines 1-D kernel factorized networks, position and channel attention, and multiscale aggregation mechanisms with a GAN model. The 1-D kernel factorized network reduces the computational cost of 2D filtering. The position and channel attention modules enhance the discriminative ability between the lesion and non-lesion feature representations in spatial and channel dimensions, respectively. A multiscale block is also used to aggregate the coarse-to-fine features of input skin images and reduce the effect of the artifacts. SLSNet is evaluated on two publicly available datasets: ISBI 2017 and the ISIC 2018. Although SLSNet has only 2.35 million parameters, the experimental results demonstrate that it achieves segmentation results on a par with the state-of-the-art skin lesion segmentation methods with an accuracy of 97.61%, and Dice and Jaccard similarity coefficients of 90.63% and 81.98%, respectively. SLSNet can run at more than 110 frames per second (FPS) in a single GTX1080Ti GPU, which is faster than well-known deep learning-based image segmentation models, such as FCN. Therefore, SLSNet can be used for practical dermoscopic applications.

Abstract Herein, the problem of vehicle re‐identification using distance comparison of images in CNN latent spaces is addressed. First, the impact of the distance metrics, comparing performances obtained with different metrics is studied: the minimal Euclidean distance ( MED ), the minimal cosine distance ( MCD ) and the residue of the sparse coding reconstruction ( RSCR ). These metrics are applied using features extracted from five different CNN architectures, namely ResNet18, AlexNet, VGG16, InceptionV3 and DenseNet201. We use the specific vehicle re‐identification dataset VeRi to fine‐tune these CNNs and evaluate results. Overall, independently of the CNN used, MCD outperforms MED , commonly used in the literature. These results are confirmed on other vehicle retrieval datasets. Second, the state‐of‐the‐art image‐to‐track process (I2TP) is extended to a track‐to‐track process (T2TP). The three distance metrics are extended to measure distance between tracks, enabling T2TP. T2TP and I2TP are compared using the same CNN models. Results show that T2TP outperforms I2TP for MCD and RSCR. T2TP combining DenseNet201 and MCD ‐based metrics exhibits the best performances, outperforming the state‐of‐the‐art I2TP‐based models. Finally, experiments highlight two main results: i) the impact of metric choice in vehicle re‐identification, and ii) T2TP improves the performances compared with I2TP, especially when coupled with MCD ‐based metrics.

Determining the links between large amounts of video data with no prior knowledge of the camera positions is a hard task to automate. From a collection of videos acquired from static cameras simultaneously, we propose a method for finding groups of videos with overlapping fields of view. Each video is first processed individually: at regular time steps, objects are detected and are assigned a category and an appearance descriptor. Next, the video is split into cells at different resolutions and we assign to each cell its story: it consists of the list of objects detected in the cell over time. Once the stories are established for each video, the links between cells of different videos are determined by comparing their stories: two cells are linked if they show simultaneous detections of objects of the same category with similar appearances. Pairs of videos with overlapping fields of view are identified using these links between cells. A link graph is finally returned, in which each node represents a video, and the edges indicate pairs of overlapping videos. The approach is evaluated on a set of 63 real videos from both public datasets and live surveillance videos, as well as on 84 synthetic videos, and shows promising results.

Estimating a depth map and, at the same time, predicting the 3D pose of an object from a single 2D color image is a very challenging task. Depth estimation is typically performed through stereo vision by following several time-consuming stages, such as epipolar geometry, rectification and matching. Alternatively, when stereo vision is not useful or applicable, depth relations can be inferred from a single image as studied in this paper. More precisely, deep learning is applied in order to solve the problem of estimating a depth map from a single image. Then, that map is used for predicting the 3D pose of the main object depicted in the image. The proposed model consists of two successive neural networks. The first network is based on a Generative Adversarial Neural network (GAN). It estimates a dense depth map from the given color image. A Convolutional Neural Network (CNN) is then used to predict the 3D pose from the generated depth map through regression. The main difficulty to jointly estimate depth maps and 3D poses using deep networks is the lack of training data with both depth and viewpoint annotations. This contribution assumes a cross-domain training procedure with 3D CAD models corresponding to objects appearing in real images in order to render depth images from different viewpoints. These rendered images are then used to guide the GAN network to learn the mapping from the image domain to the depth domain. By exploiting the dataset as a source of training data, the proposed model outperforms state-of-the-art models on the PASCAL 3D+ dataset. The code of the proposed model is publicly available at https://github.com/SaddamAbdulrhman/Depth-and-Viewpoint-Estimation/tree/master.

We present an approach for ranking a collection of videos with overlapping fields of view. The ranking depends on how they allow to visualize as best as possible, i.e. with significant details, a trajectory query drawn in one of the videos. The proposed approach decomposes each video into cells and aims at estimating a correspondence map between cells from different videos using the linear correlation between their functions of activity. These latter are obtained during a training session by detecting objects in the videos and computing the coverage rate between the objects and the cells over time. The main idea is that two areas from two different videos that systematically offer presence of objects simultaneously are very likely to correspond to each other. Then, we use the correspondence between cells to find the reformulated trajectory in the other videos. Finally, we rank the videos based on the visibility they offer. We show promising results by testing three aspects: the correspondence maps, the reformulation and the ranking.

In the context of 2D/3D registration, this paper introduces an approach that allows for matching features detected in two different modalities, photographs, and 3D models, by using a common 2D representation. More precisely, 2D images are matched with a set of depth images representing the 3D model. After introducing the concept of Curvilinear Saliency, which is related to curvature estimation, we propose a new ridge and valley detector for depth images rendered from 3D models. A variant of this detector is adapted to photographs, first by considering multi-scale features and second by integrating the focus curve principle. Finally, a registration algorithm determines the correct view of the 3D model and, thus, the pose of the photograph. This approach relies on the Histogram of Curvilinear Saliency (HCS), an adaptation of the Histogram of Oriented Gradients (HOG) to the proposed features in 2D and 3D. The presented results highlight both the quality of the features detected in terms of repeatability and the interest of the approach for registration and pose estimation.

In surveillance applications, humans and vehicles are the most important common elements studied. In consequence, detecting and matching a person or a car that appears on several videos is a key problem. Many algorithms have been introduced and nowadays, a major relative problem is to evaluate precisely and to compare these algorithms, in reference to a common ground-truth. In this paper, our goal is to introduce a new dataset for evaluating multi-view based methods. This dataset aims at paving the way for multidisciplinary approaches and applications such as 4D-scene reconstruction, object identification/tracking, audio event detection and multi-source meta-data modeling and querying. Consequently, we provide two sets of 25 synchronized videos with audio tracks, all depicting the same scene from multiple viewpoints, each set of videos following a detailed scenario consisting in comings and goings of people and cars. Every video was annotated by regularly drawing bounding boxes on every moving object with a flag indicating whether the object is fully visible or occluded, specifying its category (human or vehicle), providing visual details (for example clothes types or colors), and timestamps of its apparitions and disappearances. Audio events are also annotated by a category and timestamps.

Insect detection is one of the most challenging problems of biometric image processing. This study focuses on developing a method to detect both individual insects and touching insects from trap images in extreme conditions. This method is able to combine recent approaches on contour‐based and region‐based segmentation. More precisely, the two contributions are: an adaptive k ‐means clustering approach by using the contour's convex hull and a new region merging algorithm. Quantitative evaluations show that the proposed method can detect insects with higher accuracy than that of the most used approaches.

The advantage of skeleton-based 3D reconstruction is to completely generate a single 3D object from well chosen views. Having numerous views is necessary for a reliable reconstruction but projections of skeletons lead to different topologies. We reconstruct 3D objects with curved medial axis (whose topology is a tree) from the perspective skeletons on an arbitrary number of calibrated acquisitions. The main contribution is to estimate the 3D skeleton, from multiple images: its topology is chosen as the closest to those of the perspective skeletons on the set of images, which means that the number of topology changes to map the 3D skeleton topology to topologies on images is minimum. This way, the number of matched branches is maximum.

Object recognition based on local features computed at multiple locations is robust to occlusions, strong viewpoint changes and object deformations. These features should be repeatable, precise and distinctive. We present an operator for repeatable feature detection on depth images (relative to 3D models) as well as 2D intensity images. The proposed detector is based on estimating the curviness saliency at multiple scales in each kind of image. We also propose quality measures that evaluate the repeatability of the features between depth and intensity images. The experiments show that the proposed detector outperforms both the most powerful, classical point detectors (e.g., SIFT) and edge detection techniques.

This paper proposes a new algorithm for automatic crack detection from 2D pavement images. It strongly relies on the localization of minimal paths within each image, a path being a series of neighboring pixels and its score being the sum of their intensities. The originality of the approach stems from the proposed way to select a set of minimal paths and the two postprocessing steps introduced to improve the quality of the detection. Such an approach is a natural way to take account of both the photometric and geometric characteristics of pavement images. An intensive validation is performed on both synthetic and real images (from five different acquisition systems), with comparisons to five existing methods. The proposed algorithm provides very robust and precise results in a wide range of situations, in a fully unsupervised manner, which is beyond the current state of the art.

We present a novel approach to reconstruct a 3D object from images corresponding to two different viewpoints: we estimate the skeleton of the object instead of its surface. The originality of the method is to be able to reconstruct a complete tubular 3D object from only two input images. Unlike classical reconstruction methods like multiview stereo, this approach does not rely on interest points but estimates the topology of the object and derives its surface. Our contributions are twofold. First, given two perspective images of the 3D shape, the projection of the skeleton is computed in 2D. Second the 3D skeleton is reconstructed from the two projections using triangulation and matching. A mesh is finally derived for each skeleton branch.

This paper proposes a new algorithm for crack detection based on the selection of minimal paths. It takes account of both photometric and geometric characteristics and requires few information a priori. It is validated on synthetic and real images.

Thin-plate spline warps have been shown to be very effective as a parameterized model of the optic flow field between images of various deforming surfaces. Examples include a sheet of paper being manually handled. Recent work has used such warps for images of smooth rigid surfaces. Standard thin-plate spline warps are not rigid, in the sense that they do not satisfy the epipolar geometry constraint, and are intrinsically affine, in the sense of the affine camera model. We propose three types of warps based on the thin-plate spline. The first one is a flexible rigid warp. It describes the optic flow field induced by a smooth rigid surface, and satisfies the affine epipolar geometry constraint. The second and third ones extend the standard thin-plate spline and the proposed rigid flexible warp to the perspective camera model. The properties of these warps are studied in details, and a hierarchy is defined. Experimental results on simulated and real data are reported, showing that the proposed warps outperform the standard one in several cases of interest.

In the context of computer vision, matching can be done using correlation measures. This paper presents the classification of fifty measures into five families. In addition, eighteen new measures based on robust statistics are presented to deal with the problem of occlusions. An evaluation protocol is proposed and the results show that robust measures (one of the five families), including the new measures, give the best results near occlusions. 1