FlowNet3D++: 幾何学的損失による深層シーンフロー推定
FlowNet3D++: Geometric Losses For Deep Scene Flow Estimation
点と平面の距離やベクトル間の角度整合性といった幾何学的制約を損失に組み込み、シーンフロー推定の精度を57.85%から63.43%に向上させた。さらに動的3D再構成のベンチマークを提案し、再構成誤差を最大15%削減した。
著者: Zirui Wang, Shuda Li, Henry Howard-Jenkins, Victor Adrian Prisacariu, Min Chen
分類: cs.CV, cs.RO
原文アブストラクト
We present FlowNet3D++, a deep scene flow estimation network. Inspired by classical methods, FlowNet3D++ incorporates geometric constraints in the form of point-to-plane distance and angular alignment between individual vectors in the flow field, into FlowNet3D. We demonstrate that the addition of these geometric loss terms improves the previous state-of-art FlowNet3D accuracy from 57.85% to 63.43%. To further demonstrate the effectiveness of our geometric constraints, we propose a benchmark for flow estimation on the task of dynamic 3D reconstruction, thus providing a more holistic and practical measure of performance than the breakdown of individual metrics previously used to evaluate scene flow. This is made possible through the contribution of a novel pipeline to integrate point-based scene flow predictions into a global dense volume. FlowNet3D++ achieves up to a 15.0% reduction in reconstruction error over FlowNet3D, and up to a 35.2% improvement over KillingFusion alone. We will release our scene flow estimation code later.
