Sorry, you need to enable JavaScript to visit this website.

Semantic Scene Completion (SSC) aims to jointly predict semantic categories and 3D occupancy of a scene from coarse inputs, which is crucial for providing reliable perception in autonomous driving. In this paper, we enhance existing SSC models by unveiling the vanishing point region, specifically addressing challenges posed by tiny objects and voxels distant from the monocular camera. At the core of our method, we propose the Vanishing Point Aggregator (VPA) to prioritize features in high-density central areas.

Fixed pattern noise (FPN) is a temporally constant noise present on videos due to the non-uniformities of the sensors that may exhibit spatial correlation, typically across columns and/or rows.
Acquiring real clean/noisy data is particularly challenging in the case of FPN, leading supervised FPN denoising networks to train using generated data.
Self-supervised approaches for denoising allow training directly on real noisy sequences, avoiding the biases introduced by synthetic data.

Video supplementary materials

GEOSCALER: GEOMETRY AND RENDERING-AWARE DOWNSAMPLING OF 3D MESH TEXTURES: High-resolution texture maps are necessary to accurately represent real-world objects with 3D meshes. The large sizes of textures can bottleneck the real-time rendering of high-quality virtual 3D scenes on devices that have low computational budgets and limited memory. Downsampling the texture maps directly addresses the issue, albeit at the cost of visual fidelity. Traditionally, downsampling of texture maps is performed using methods such as bicubic interpolation and the Lanczos algorithm.

Point cloud compression is a critical component in 3D vision systems utilizing point cloud data to represent the physical world. Existing works on point cloud compression separately tackle Octree-based and Feature-based coding of point clouds despite their underlying similarities. In this work, we present UH-PCC, a unified hierarchical model for point cloud compression, that synergizes Octree-based and Feature-based coding under a single model with shared parameters via a Hierarchical Feature Coding (HFC) framework.

This is supplementary to our paper titled: LEVERAGING 3D GAUSSIAN SPLATTING TO ENHANCE FACE PARSING

This document is the supplementary material of our submitted paper: MS-RAFT-3D: A Multi-Scale Architecture for Recurrent Image-based Scene Flow.

Recent advancements in text-to-image (T2I) have improved synthesis results, but challenges remain in layout control and generating omnidirectional panoramic images. Dense T2I (DT2I) and spherical T2I (ST2I) models address these issues, but so far no unified approach exists. Trivial approaches, like prompting a DT2I model to generate panoramas can not generate proper spherical distortions and seamless transitions at the borders. Our work shows that spherical dense text-to-image (SDT2I) can be achieved by integrating training-free DT2I approaches into finetuned panorama models.

In this supplementary material, we provide implementation details of our method, training details of our method, additional experiments and visual results.

Testing a post