Recovering and Simulating Pedestrians in the Wild
- URL: http://arxiv.org/abs/2011.08106v1
- Date: Mon, 16 Nov 2020 17:16:32 GMT
- Title: Recovering and Simulating Pedestrians in the Wild
- Authors: Ze Yang, Siva Manivasagam, Ming Liang, Bin Yang, Wei-Chiu Ma, Raquel
Urtasun
- Abstract summary: We propose to recover the shape and motion of pedestrians from sensor readings captured in the wild by a self-driving car driving around.
We incorporate the reconstructed pedestrian assets bank in a realistic 3D simulation system.
We show that the simulated LiDAR data can be used to significantly reduce the amount of real-world data required for visual perception tasks.
- Score: 81.38135735146015
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Sensor simulation is a key component for testing the performance of
self-driving vehicles and for data augmentation to better train perception
systems. Typical approaches rely on artists to create both 3D assets and their
animations to generate a new scenario. This, however, does not scale. In
contrast, we propose to recover the shape and motion of pedestrians from sensor
readings captured in the wild by a self-driving car driving around. Towards
this goal, we formulate the problem as energy minimization in a deep structured
model that exploits human shape priors, reprojection consistency with 2D poses
extracted from images, and a ray-caster that encourages the reconstructed mesh
to agree with the LiDAR readings. Importantly, we do not require any
ground-truth 3D scans or 3D pose annotations. We then incorporate the
reconstructed pedestrian assets bank in a realistic LiDAR simulation system by
performing motion retargeting, and show that the simulated LiDAR data can be
used to significantly reduce the amount of annotated real-world data required
for visual perception tasks.
Related papers
- Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles [81.29018359825872]
This paper consolidates a set of good practices to finetune large pretrained models for a real-world task.
Specifically, we develop several strategies to account for discrepancies between the synthetic data and real driving data.
Our insights lead to effective finetuning that results in a $68.8%$ reduction in FID for novel view synthesis over prior arts.
arXiv Detail & Related papers (2024-12-19T03:39:13Z) - Street Gaussians without 3D Object Tracker [86.62329193275916]
Existing methods rely on labor-intensive manual labeling of object poses to reconstruct dynamic objects in canonical space and move them based on these poses during rendering.
We propose a stable object tracking module by leveraging associations from 2D deep trackers within a 3D object fusion strategy.
We address inevitable tracking errors by further introducing a motion learning strategy in an implicit feature space that autonomously corrects trajectory errors and recovers missed detections.
arXiv Detail & Related papers (2024-12-07T05:49:42Z) - Neural Rendering based Urban Scene Reconstruction for Autonomous Driving [8.007494499012624]
We propose a multimodal 3D scene reconstruction using a framework combining neural implicit surfaces and radiance fields.
Dense 3D reconstruction has many applications in automated driving including automated annotation validation.
We demonstrate qualitative and quantitative results on challenging automotive scenes.
arXiv Detail & Related papers (2024-02-09T23:20:23Z) - CADSim: Robust and Scalable in-the-wild 3D Reconstruction for
Controllable Sensor Simulation [44.83732884335725]
Sensor simulation involves modeling traffic participants, such as vehicles, with high quality appearance and articulated geometry.
Current reconstruction approaches struggle on in-the-wild sensor data, due to its sparsity and noise.
We present CADSim, which combines part-aware object-class priors via a small set of CAD models with differentiable rendering to automatically reconstruct vehicle geometry.
arXiv Detail & Related papers (2023-11-02T17:56:59Z) - Aerial Monocular 3D Object Detection [67.20369963664314]
DVDET is proposed to achieve aerial monocular 3D object detection in both the 2D image space and the 3D physical space.
To address the severe view deformation issue, we propose a novel trainable geo-deformable transformation module.
To encourage more researchers to investigate this area, we will release the dataset and related code.
arXiv Detail & Related papers (2022-08-08T08:32:56Z) - LiDARsim: Realistic LiDAR Simulation by Leveraging the Real World [84.57894492587053]
We develop a novel simulator that captures both the power of physics-based and learning-based simulation.
We first utilize ray casting over the 3D scene and then use a deep neural network to produce deviations from the physics-based simulation.
We showcase LiDARsim's usefulness for perception algorithms-testing on long-tail events and end-to-end closed-loop evaluation on safety-critical scenarios.
arXiv Detail & Related papers (2020-06-16T17:44:35Z) - SurfelGAN: Synthesizing Realistic Sensor Data for Autonomous Driving [27.948417322786575]
We present a simple yet effective approach to generate realistic scenario sensor data.
Our approach uses texture-mapped surfels to efficiently reconstruct the scene from an initial vehicle pass or set of passes.
We then leverage a SurfelGAN network to reconstruct realistic camera images for novel positions and orientations of the self-driving vehicle.
arXiv Detail & Related papers (2020-05-08T04:01:14Z) - Transferable Active Grasping and Real Embodied Dataset [48.887567134129306]
We show how to search for feasible viewpoints for grasping by the use of hand-mounted RGB-D cameras.
A practical 3-stage transferable active grasping pipeline is developed, that is adaptive to unseen clutter scenes.
In our pipeline, we propose a novel mask-guided reward to overcome the sparse reward issue in grasping and ensure category-irrelevant behavior.
arXiv Detail & Related papers (2020-04-28T08:15:35Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.