オブジェクト合成可能なNeRFによる視覚のSim-to-Realギャップ解消
Closing the Visual Sim-to-Real Gap with Object-Composable NeRFs
実画像から物体を抽出して新しいシーンに合成できるNeRFモデルCOV-NeRFを提案し、写実的な描画と多様な2D/3D教師信号を生成して視覚タスクのsim-to-realギャップを迅速に埋める。
著者: Nikhil Mishra, Maximilian Sieb, Pieter Abbeel, Xi Chen
分類: cs.RO, cs.CV, cs.LG
原文アブストラクト
Deep learning methods for perception are the cornerstone of many robotic systems. Despite their potential for impressive performance, obtaining real-world training data is expensive, and can be impractically difficult for some tasks. Sim-to-real transfer with domain randomization offers a potential workaround, but often requires extensive manual tuning and results in models that are brittle to distribution shift between sim and real. In this work, we introduce Composable Object Volume NeRF (COV-NeRF), an object-composable NeRF model that is the centerpiece of a real-to-sim pipeline for synthesizing training data targeted to scenes and objects from the real world. COV-NeRF extracts objects from real images and composes them into new scenes, generating photorealistic renderings and many types of 2D and 3D supervision, including depth maps, segmentation masks, and meshes. We show that COV-NeRF matches the rendering quality of modern NeRF methods, and can be used to rapidly close the sim-to-real gap across a variety of perceptual modalities.
