The field of 3D reconstruction and rendering is rapidly advancing, with a focus on improving the accuracy and efficiency of reconstruction methods. Recent developments have seen the introduction of new techniques, such as Gaussian splatting and neural implicit surface reconstruction, which are able to handle complex scenes and provide high-quality reconstructions. Additionally, there is a growing interest in applying these techniques to real-world applications, such as autonomous driving and underwater robotics. Noteworthy papers in this area include QuickSplat, which accelerates 3D surface reconstruction by learning data-driven priors, and Geometric Prior-Guided Neural Implicit Surface Reconstruction, which applies multiple geometric constraints to improve the accuracy of reconstructions. Overall, the field is moving towards more efficient and accurate reconstruction methods, with a focus on practical applications.
Advances in 3D Reconstruction and Rendering
Sources
QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization
RefRef: A Synthetic Dataset and Benchmark for Reconstructing Refractive and Reflective Objects
VIN-NBV: A View Introspection Network for Next-Best-View Selection for Resource-Efficient 3D Reconstruction
LMLCC-Net: A Semi-Supervised Deep Learning Model for Lung Nodule Malignancy Prediction from CT Scans using a Novel Hounsfield Unit-Based Intensity Filtering
Gaussian Wave Splatting for Computer-Generated Holography
Geometric Prior-Guided Neural Implicit Surface Reconstruction in the Wild
ACT-R: Adaptive Camera Trajectories for 3D Reconstruction from Single Image
FOCI: Trajectory Optimization on Gaussian Splats
TUGS: Physics-based Compact Representation of Underwater Scenes by Tensorized Gaussian
Template-Guided Reconstruction of Pulmonary Segments with Neural Implicit Functions
FreeDriveRF: Monocular RGB Dynamic NeRF without Poses for Autonomous Driving via Point-Level Dynamic-Static Decoupling
Sparse Point Cloud Patches Rendering via Splitting 2D Gaussians
Large-Scale Gaussian Splatting SLAM
High Quality Underwater Image Compression with Adaptive Correction and Codebook-based Augmentation
Advances in Radiance Field for Dynamic Scene: From Neural Field to Gaussian Field