単眼セマンティック占有グリッドマッピングのための畳み込み変分エンコーダ・デコーダネットワーク
Monocular Semantic Occupancy Grid Mapping with Convolutional Variational Encoder-Decoder Networks
単眼画像から車載視点の意味的・距離的な占有グリッドを変分エンコーダ・デコーダで直接推定し、平面仮定の従来法より高精度かつ車両動揺に頑健でリアルタイム動作することを示した論文。
著者: Chenyang Lu, Marinus Jacobus Gerardus van de Molengraft, Gijs Dubbelman
分類: cs.RO, cs.CV
原文アブストラクト
In this work, we research and evaluate end-to-end learning of monocular semantic-metric occupancy grid mapping from weak binocular ground truth. The network learns to predict four classes, as well as a camera to bird's eye view mapping. At the core, it utilizes a variational encoder-decoder network that encodes the front-view visual information of the driving scene and subsequently decodes it into a 2-D top-view Cartesian coordinate system. The evaluations on Cityscapes show that the end-to-end learning of semantic-metric occupancy grids outperforms the deterministic mapping approach with flat-plane assumption by more than 12% mean IoU. Furthermore, we show that the variational sampling with a relatively small embedding vector brings robustness against vehicle dynamic perturbations, and generalizability for unseen KITTI data. Our network achieves real-time inference rates of approx. 35 Hz for an input image with a resolution of 256x512 pixels and an output map with 64x64 occupancy grid cells using a Titan V GPU.
