BlitzNet: A Real-Time Deep Network for Scene Understanding

Nikita Dvornik; Konstantin Shmelkov; Julien Mairal; Cordelia Schmid

doi:10.1109/ICCV.2017.447

BlitzNet: A Real-Time Deep Network for Scene Understanding

Nikita Dvornik ¹ Konstantin Shmelkov ¹ Julien Mairal ¹ Cordelia Schmid ¹

1 Thoth - Apprentissage de modèles à partir de données massives

LJK - Laboratoire Jean Kuntzmann, Inria Grenoble - Rhône-Alpes

Abstract : Real-time scene understanding has become crucial in many applications such as autonomous driving. In this paper, we propose a deep architecture, called BlitzNet, that jointly performs object detection and semantic segmentation in one forward pass, allowing real-time computations. Besides the computational gain of having a single network to perform several tasks, we show that object detection and semantic segmentation benefit from each other in terms of accuracy. Experimental results for VOC and COCO datasets show state-of-the-art performance for object detection and segmentation among real time systems.

Keywords : Object detection Real-time Neural Networks Semantic Segmentation

Document type :

Conference papers

Domain :

Computer Science [cs]

Liste complète des métadonnées

Cited literature [27 references] Display Hide Download

https://hal.archives-ouvertes.fr/hal-01573361
Contributor : Nikita Dvornik <>
Submitted on : Wednesday, August 9, 2017 - 11:43:46 AM
Last modification on : Thursday, September 13, 2018 - 11:32:08 AM

egpaper_final.pdf

Files produced by the author(s)

BlitzNet: A Real-Time Deep Network for Scene Understanding

File

Identifiers

Collections

Citation

Export

Metrics

822

5025

Contact

BlitzNet: A Real-Time Deep Network for Scene Understanding

File

Identifiers

Collections

Citation

Export

Share

Metrics

822

5025

Contact