AI Paper Atlas

Research map / 3D vision / 3D generation

CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting

Jiezhi Yang, Khushi P Desai, Charles Packer, Harshil bhatia, Nicholas Rhinehart, Rowan McAllister, Joseph E Gonzalez

ECCV 2024 · 3D vision › 3D generation · approach: Monocular 3D object reconstruction · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
sample n=10 latents from the mixture density network in the controller (versus 2 or 35) (ablation)CARFF controller · Multi-Scene Intersection planningsuccess rate (30 trials): 17/30 and 19/30 -> 30/30 on the hard caseImproved
use probabilistic CARFF controller instead of overconfident or underconfident controllerCARFF controller · Multi-Scene Two Lane Merge planningsuccess rate (30 trials): 0/30 -> 30/30Improved
add convolutional layer to the ViT-based PC-VAE encoder (ablation)PC-VAE · Multi-Scene Two Lane MergeSVM accuracy (%): 29.67 -> 88.33Improved
add convolutional layer to the ViT-based PC-VAE encoder (ablation)PC-VAE · Multi-Scene Approaching IntersectionSVM accuracy (%): 83.83 -> 89.17Improved
freeze pretrained ViT encoder (ablation)PC-VAE · Multi-Scene Approaching IntersectionSVM accuracy (%): 29.83 -> 89.17Improved
use pretrained ViT encoder instead of MobileNet (ablation)PC-VAE · Multi-Scene Approaching Intersectiontrain PSNR: 19.37 -> 26.47Improved
add pose conditioning to the VAE decoder (ablation)VAE · Multi-Scene Approaching Intersectionnovel-view PSNR: 9.84 -> 26.37Improved
use CARFF's PC-VAE instead of NeRF-VAENeRF-VAE · CARLA datasetstraining stabilityImproved
increase KL weight with delayed linear schedule (ablation)PC-VAE · CARLA datasetslatent qualityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy