Multi-Category Mesh Reconstruction From Image Collections

Simoni, Alessandro; Pini, Stefano; Vezzani, Roberto; Cucchiara, Rita

doi:10.1109/3dv53792.2021.00139

Cited by 7 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This neural model is trained over the same set of photorealistic RGB images of the considered categories and learns how to infer the 3D mesh of unseen instances from a single viewpoint. In this work a box-supervised 3D model reconstructor is developed on top of Multi-Category Mesh Reconstruction (MCMR) [26] as in Fig. 4.…”

Section: D Model Reconstructionmentioning

confidence: 99%

See 1 more Smart Citation

i2c-net: Using Instance-Level Neural Networks for Monocular Category-Level 6D Pose Estimation

Remus

D’Avella

Felice

et al. 2023

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Object detection and pose estimation are strict requirements for many robotic grasping and manipulation applications to endow robots with the ability to grasp objects with different properties in cluttered scenes and with various lighting conditions. This work proposes the framework i2c-net to extract the 6D pose of multiple objects belonging to different categories, starting from an instance-level pose estimation network and relying only on RGB images. The network is trained on a custom-made synthetic photo-realistic dataset, generated from some base CAD models, opportunely deformed, and enriched with real textures for domain randomization purposes. At inference time, the instance-level network is employed in combination with a 3D mesh reconstruction module, achieving category-level capabilities. Depth information is used for post-processing as a correction. Tests conducted on real objects of the YCB-V and NOCS-REAL datasets outline the high accuracy of the proposed approach.

show abstract

Section: D Model Reconstructionmentioning

confidence: 99%

“…It is possible to learn multiple meanshapes and let a classifier select the most suitable one with respect to the image features. In the presented approach, the loss function in [26] is completed with the 3D bounding box supervision by introducing the term…”

Section: D Model Reconstructionmentioning

confidence: 99%

i2c-net: Using Instance-Level Neural Networks for Monocular Category-Level 6D Pose Estimation

Remus

D’Avella

Felice

et al. 2023

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Pixel2Mesh [26] treats a mesh as a graph and applies graph convolution [27] for vertex feature extraction and graph unpooling to subdivide the mesh for detailed refinement. Using differentiable mesh rendering [28], [29], the 3D mesh structure of an object can be learned from 2D images [30]- [32]. Mesh R-CNN [33] simultaneously detects objects and reconstructs their 3D mesh shape.…”

Section: Related Workmentioning

confidence: 99%

TerrainMesh: Metric-Semantic Terrain Reconstruction from Aerial Images Using Joint 2D-3D Learning

Feng¹,

Atanasov²

2022

Preprint

View full text Add to dashboard Cite

This paper considers outdoor terrain mapping using RGB images obtained from an aerial vehicle. While feature-based localization and mapping techniques deliver real-time vehicle odometry and sparse keypoint depth reconstruction, a dense model of the environment geometry and semantics (vegetation, buildings, etc.) is usually recovered offline with significant computation and storage. This paper develops a joint 2D-3D learning approach to reconstruct a local metric-semantic mesh at each camera keyframe maintained by a visual odometry algorithm. Given the estimated camera trajectory, the local meshes can be assembled into a global environment model to capture the terrain topology and semantics during online operation. A local mesh is reconstructed using an initialization and refinement stage. In the initialization stage, we estimate the mesh vertex elevation by solving a least squares problem relating the vertex barycentric coordinates to the sparse keypoint depth measurements. In the refinement stage, we associate 2D image and semantic features with the 3D mesh vertices using camera projection and apply graph convolution to refine the mesh vertex spatial coordinates and semantic features based on joint 2D and 3D supervision. Quantitative and qualitative evaluation using real aerial images show the potential of our method to support environmental monitoring and surveillance applications.

show abstract