Corentin Dumery

Visual object counting is a fundamental computer vision task in industrial inspection, where accurate, high-throughput inventory tracking and quality assurance are critical. Moreover, manufactured parts are often too light to reliably deduce their count from their weight, or too heavy to move the stack on a scale safely and practically, making automated visual counting the more robust solution in many scenarios. However, existing methods struggle with stacked 3D items in containers, pallets, or bins, where most objects are heavily occluded and only a few are directly visible. To address this important yet underexplored challenge, we propose a novel 3D counting approach that decomposes the task into two complementary subproblems: estimating the 3D geometry of the stack and its occupancy ratio from multi-view images. By combining geometric reconstruction with deep learning-based depth analysis, our method can accurately count identical manufactured parts inside containers, even when they are irregularly stacked and partially hidden. We validate our 3D counting pipeline on large-scale synthetic and diverse real-world data with manually verified total counts, demonstrating robust performance under realistic inspection conditions.

Sewing patterns define the structural foundation of garments and are essential for applications such as fashion design, fabrication, and physical simulation. Despite progress in automated pattern generation, accurately modeling sewing patterns remains difficult due to the broad variability in panel geometry and seam arrangements. In this work, we introduce a sewing pattern modeling method based on an implicit representation. We represent each panel using a signed distance field that defines its boundary and an unsigned distance field that identifies seam endpoints, and encode these fields into a continuous latent space that enables differentiable meshing. A latent flow matching model learns distributions over panel combinations in this representation, and a stitching prediction module recovers seam relations from extracted edge segments. This formulation allows accurate modeling and generation of sewing patterns with complex structures. We further show that it can be used to estimate sewing patterns from images with improved accuracy relative to existing approaches, and supports applications such as pattern completion and refitting, providing a practical tool for digital fashion design.

Neural Radiance Fields (NeRF) has emerged as a compelling framework for scene representation and 3D recovery. To improve its performance on real-world data, depth regularizations have proven to be the most effective ones. However, depth estimation models not only require expensive 3D supervision in training, but also suffer from generalization issues. As a result, the depth estimations can be erroneous in practice, especially for outdoor unbounded scenes. In this paper, we propose to employ view-consistent distributions instead of fixed depth value estimations to regularize NeRF training. Specifically, the distribution is computed by utilizing both low-level color features and high-level distilled features from foundation models at the projected 2D pixel-locations from per-ray sampled 3D points. By sampling from the view-consistency distributions, an implicit regularization is imposed on the training of NeRF. We also utilize a depth-pushing loss that works in conjunction with the sampling technique to jointly provide effective regularizations for eliminating the failure modes. Extensive experiments conducted on various scenes from public datasets demonstrate that our proposed method can generate significantly better novel view synthesis results than state-of-the-art NeRF variants as well as different depth regularization methods.

3D Gaussian Splatting (3DGS) has become increasingly popular in 3D scene reconstruction for its high visual accuracy. However, uncertainty estimation of 3DGS scenes remains underexplored and is crucial to downstream tasks such as asset extraction and scene completion. Since the appearance of 3D gaussians is view-dependent, the color of a gaussian can thus be certain from an angle and uncertain from another. We thus propose to model uncertainty in 3DGS as an additional view-dependent per-gaussian feature that can be modeled with spherical harmonics. This simple yet effective modeling is easily interpretable and can be integrated into the traditional 3DGS pipeline. It is also significantly faster than ensemble methods while maintaining high accuracy, as demonstrated in our experiments.

Polycube-maps are used as base-complexes in various fields of computational geometry, including the generation of regular all-hexahedral meshes free of internal singularities. However, the strict alignment constraints behind polycube-based methods make their computation challenging for CAD models used in numerical simulation via Finite Element Method (FEM). We propose a novel approach based on an evolutionary algorithm to robustly compute polycube-maps in this context. We address the labeling problem, which aims to precompute polycube alignment by assigning one of the base axes to each boundary face on the input. Previous research has described ways to initialize and improve a labeling via greedy local fixes. However, such algorithms lack robustness and often converge to inaccurate solutions for complex geometries. Our proposed framework alleviates this issue by embedding labeling operations in an evolutionary heuristic, defining fitness, crossover, and mutations in the context of labeling optimization. We evaluate our method on a thousand smooth and CAD meshes, showing Evocube converges to valid labelings on a wide range of shapes. The limitations of our method are also discussed thoroughly.

Polycube‐maps are used as base‐complexes in various fields of computational geometry, including the generation of regular all‐hexahedral meshes free of internal singularities. However, the strict alignment constraints behind polycube‐based methods make their computation challenging for CAD models used in numerical simulation via finite element method (FEM). We propose a novel approach based on an evolutionary algorithm to robustly compute polycube‐maps in this context.We address the labelling problem, which aims to precompute polycube alignment by assigning one of the base axes to each boundary face on the input. Previous research has described ways to initialize and improve a labelling via greedy local fixes. However, such algorithms lack robustness and often converge to inaccurate solutions for complex geometries. Our proposed framework alleviates this issue by embedding labelling operations in an evolutionary heuristic, defining fitness, crossover, and mutations in the context of labelling optimization. We evaluate our method on a thousand smooth and CAD meshes, showing Evocube converges to accurate labellings on a wide range of shapes. The limitations of our method are also discussed thoroughly.