AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Neural-Sim: Learning to Generate Training Data with NeRF

Yunhao Ge, Harkirat Behl, Jiashu Xu, Suriya Gunasekar, Neel Joshi, Yale Song, Xin Wang, Laurent Itti, Vibhav Vineet

ECCV 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Cross-domain translation and pose · 27 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
add bi-level optimization of rendering parameters (NSO vs NS)Neural-Sim · YCB-synthetic (21 objects)mAP: 68.4 -> 82.1Improved
use NSO instead of Auto-SimAuto-Sim · YCB-synthetic (21 objects)mAP: 69.3 -> 82.1Improved
add NSO to optimize pose distribution when train/test pose gap existsNeural-Sim · YCB-synthetic pitcher and drilleraverage precision: almost 30% and 10%Improved
use differentiable NSO instead of REINFORCE-based LTS/Auto-Sim for poseLTS and Auto-Sim · YCB-synthetic pitcheraverage precision: almost 34% and 11%Improved
add NSO illumination optimization with NeRF-wNeural-Sim · YCB-synthetic illuminationaverage precision: 16% and 15%Improved
add NSO on real YCB-in-the-wild imagesNeural-Sim · YCB-in-the-wild pitcheraccuracy: almost 60%Improved
add optimization (NSO) over NS in few-shot YCB-Videotrain(pre)+ours · YCB-Video 10-shot and 25-shotmAP: 3.9 -> 4.9; 4.6 -> 4.9Improved
add NS synthetic data to few-shot real dataOnly YCBV-train · YCB-Video few-shotmAP: 0.45 -> 3.9; 0.49 -> 4.6Improved
add NSO optimization to NS in zero-shottrain(pre)+ours · YCB-Video 0-shotmAP: 2.3 -> 4.5Improved
add optimization to combined real+synthetic trainingtrain(com)+ours · YCB-Video limited datamAP: 36.7 -> 37.4; 43.6 -> 44.9; 47.9 -> 48.9Improved
use NSO instead of Auto-Sim on combined limited dataAuto-Sim (com) · YCB-Video limited data (10%)mAP: 48.3 -> 48.9Improved
gap between train and test pose or zoom distributionsobject detector · YCB-synthetic pitcher and drilleraverage precision: almost 30% and 10% dropHurt

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