Scene flow is a challenging task aimed at jointly estimating the 3D structure and motion of the sensed environment. Although deep learning solutions achieve outstanding performance in terms of accuracy, these approaches divide the whole problem into standalone tasks (stereo and optical flow) addressing them with independent networks. Such a strategy dramatically increases the complexity of the training procedure and requires power-hungry GPUs to infer scene flow barely at 1 FPS. Conversely, we propose DWARF, a novel and lightweight architecture able to infer full scene flow jointly reasoning about depth and optical flow easily and elegantly trainable end-to-end from scratch. Moreover, since ground truth images for full scene flow are scarce, we propose to leverage on the knowledge learned by networks specialized in stereo or flow, for which much more data are available, to distill proxy annotations. Exhaustive experiments show that i) DWARF runs at about 10 FPS on a single high-end GPU and about 1 FPS on NVIDIA Jetson TX2 embedded at KITTI resolution, with moderate drop in accuracy compared to 10× deeper models, ii) learning from many distilled samples is more effective than from the few, annotated ones available. Code available at: https://github.com/FilippoAleotti/Dwarf-Tensorflow

F. Aleotti, M.P. (2020). Learning end-to-end scene flow by distilling single tasks knowledge. Palo Alto, California USA : AAAI Press [10.1609/aaai.v34i07.6613].

Learning end-to-end scene flow by distilling single tasks knowledge

F. Aleotti;M. Poggi;F. Tosi;S. Mattoccia
2020

Abstract

Scene flow is a challenging task aimed at jointly estimating the 3D structure and motion of the sensed environment. Although deep learning solutions achieve outstanding performance in terms of accuracy, these approaches divide the whole problem into standalone tasks (stereo and optical flow) addressing them with independent networks. Such a strategy dramatically increases the complexity of the training procedure and requires power-hungry GPUs to infer scene flow barely at 1 FPS. Conversely, we propose DWARF, a novel and lightweight architecture able to infer full scene flow jointly reasoning about depth and optical flow easily and elegantly trainable end-to-end from scratch. Moreover, since ground truth images for full scene flow are scarce, we propose to leverage on the knowledge learned by networks specialized in stereo or flow, for which much more data are available, to distill proxy annotations. Exhaustive experiments show that i) DWARF runs at about 10 FPS on a single high-end GPU and about 1 FPS on NVIDIA Jetson TX2 embedded at KITTI resolution, with moderate drop in accuracy compared to 10× deeper models, ii) learning from many distilled samples is more effective than from the few, annotated ones available. Code available at: https://github.com/FilippoAleotti/Dwarf-Tensorflow
2020
Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20)
10435
10442
F. Aleotti, M.P. (2020). Learning end-to-end scene flow by distilling single tasks knowledge. Palo Alto, California USA : AAAI Press [10.1609/aaai.v34i07.6613].
F. Aleotti, M. Poggi, F. Tosi, S. Mattoccia
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/764247
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 26
  • ???jsp.display-item.citation.isi??? 17
social impact