DeepAI AI Chat
Log In Sign Up

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

by   Qiaojun Feng, et al.

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.


page 1

page 4

page 5

page 7

page 8

page 9

page 10

page 11


Mesh Reconstruction from Aerial Images for Outdoor Terrain Mapping Using Joint 2D-3D Learning

This paper addresses outdoor terrain mapping using overhead images obtai...

Dual Grid Net: hand mesh vertex regression from single depth maps

We present a method for recovering the dense 3D surface of the hand by r...

Real-time dense 3D Reconstruction from monocular video data captured by low-cost UAVs

Real-time 3D reconstruction enables fast dense mapping of the environmen...

Conformal marked bisection for local refinement of n-dimensional unstructured simplicial meshes

We present an n-dimensional marked bisection method for unstructured con...

Connectivity-preserving Geometry Images

We propose connectivity-preserving geometry images (CGIMs), which map a ...

Multi-initialization Optimization Network for Accurate 3D Human Pose and Shape Estimation

3D human pose and shape recovery from a monocular RGB image is a challen...

Sparse Depth Completion with Semantic Mesh Deformation Optimization

Sparse depth measurements are widely available in many applications such...