Vision-Based Offline-Online Perception Paradigm for Autonomous Driving

By Germán Ros, Sebastian Ramos, Manuel Granados, Amir-Hossein Bakhtiary, David Vázquez, A. L. Peña. 2015

130 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, methodology): “Out of the previously discussed datasets, only Cityscapes and KITTI provide instance-level annotations for humans and vehicles.”
Multi-Scale Context Aggregation by Dilated Convolutions (2015, background): “As new sources of data become available, future architectures may be trained densely end-to-end, removing the need for pre-training on image classiﬁcation datasets.”
Playing for Data: Ground Truth from Computer Games (2016, methodology): “We have also performed an evaluation on the KITTI semantic segmentation dataset.”
The SYNTHIA Dataset: A Large Collection of Synthetic Images for Semantic Segmentation of Urban Scenes (2016, background): “For instance, for some ADAS and for AD, semantic segmentation is a key issue [33] [18] [29] and it requires pixel-level annotations (i.e. obtained by delineating the silhouette of the different classes in urban scenarios, namely pedestrian, vehicle, road, sidewalk , vegetation, building, etc).”
SegFlow: Joint Learning for Video Object Segmentation and Optical Flow (2017, background): “Video analysis has attracted much attention in recent years due to the numerous vision applications, such as autonomous driving [15, 9, 33], video surveillance [40, 10, 20] and virtual reality [1].”
The Cityscapes Dataset (2015, background): “, [7, 18, 22]) and more challenging datasets have been created to push forward the development of these approaches.”
SMSnet: Semantic motion segmentation using deep convolutional neural networks (2017, unknown): “As described in Section IV, the KITTI benchmark does not provide any official ground truth for semantic segmentation, therefore to train the semantic stream of our network, we combine the Cityscapes dataset with the KITTI semantic ground truth from [6] and [24] to obtain the most generalized trainin”
EfficientPS: Efficient Panoptic Segmentation (2020, methodology): “In order to create panoptic annotations, we gathered semantic annotations from community driven extensions of KITTI (Xu et al, 2016; Ros et al, 2015) and combined them with the 200 training images from the KITTI semantic training set.”
Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotations (2018, methodology): “In our experiments, we consider two datasets, NYUDv240 [2], [3] and KITTI [47], [48], representing indoor and outdoor settings, respectively, and both being used extensively in the robotics community.”
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Robust Semantic Pixel-Wise Labelling (2015, background): “Table 3 shows the SegNet(R) is competitive even when temporal cues [29] are not used.”
A Review on Deep Learning Techniques Applied to Semantic Segmentation (2017, unknown): “[38] labeled 170 training images and 46 testing images (from the visual odometry challenge) with 11 classes: building, tree, sky, car, sign, road, pedestrian, fence, pole, sidewalk, and bicyclist.”
Multi-scale object candidates for generic object tracking in street scenes (2016, methodology): “[22] pre-compute a high-quality semantic map of the static parts of a scene in order to later on label the environment based on the current location within that map.”
A survey on deep learning techniques for image and video semantic segmentation (2018, unknown): “[41] labeled 170 training images and 46 testing images (from the visual odometry challenge) with 11 classes: building, tree, sky, car, sign, road, pedestrian, fence, pole, sidewalk, and bicyclist.”