ETH Zürich

By Bastian Leibe, Nico Cornelis, K. Cornelis, Luc Van Gool. 2003

How researchers see this paper

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

In Software Engineering

Citations in the corpus

RGB-D datasets using microsoft kinect or similar sensors: a survey (2016, background): “Compared to most works in which the scenes are in a very limited domain [55], this dataset is collected from much wider domains (the background is changing from one to another), facilitating multiple applications.”
The Mapillary Vistas Dataset for Semantic Understanding of Street Scenes (2017, methodology): “To gain deeper understanding of the complex interactions between e.g. traffic participants in street-level images, a lot of research efforts and investments have recently gone into designing and creating datasets which were used to train deep models for this specific task, such as CamVid [3], the KI”
Contextually Guided Semantic Labeling and Search for 3D Point Clouds (2011, methodology): “In some recent works, 3D layout or depths have been used for improving object detection (e.g., Saxena et al., 2005; Hoiem et al., 2006; Saxena et al., 2008, 2009; Hedau et al., 2010; Lee et al., 2010; Leibe et al., 2007; Heitz et al., 2008; Li et al., 2012; Batra and Saxena, 2012).”
Indoor scene segmentation using a structured light sensor (2011, methodology): “[12] use a car-mounted stereo rig to reason about the depth of a scene, detect pedestrians and cars, and track them over time.”
Building a database of 3D scenes from user annotations (2009, methodology): “Also relevant are methods to recover geometric information from image information [27, 28, 14, 24, 34, 18, 9, 21, 37].”
Alternating Decision Forests (2013, methodology): “We compare with standard Hough Forests [16] (HFs), but also with 0 0.5 1 0 0.2 0.4 0.6 0.8 1 Recall Precision HF: AP = 0.661 BHF: AP = 0.695 ADHF: AP = 0.693 (a) 0 0.5 1 0 0.2 0.4 0.6 0.8 1 Recall Precision HF: AP = 0.834 BHF: AP = 0.856 ADHF: AP = 0.857 (b) Figure 4: Recall curves of ADHF, BHF, and”
A Dynamic Conditional Random Field Model for Joint Labeling of Object and Scene Classes (2008, unknown): “[18] employ a stereo camera system together with a structure-from-motion approach to detect pedestrians and cars in urban environments.”
A mobile vision system for robust multi-person tracking (2008, background): “For applications on moving vehicles, it has been shown that visual odometry can considerably help tracking, allowing it to operate in 3D world coordinates [ 15 ].”
Semantic Labeling of 3D Point Clouds for Indoor Scenes (2011, unknown): “In some recent works, 3D layout or depths have been used for improving object detection (e.g., [11, 12, 14, 20, 21, 22, 27, 28]).”
LabelMe: Online Image Annotation and Applications (2010, methodology): “Finally, there are oth er recent relevant methods to recover geometric information f or images [21], [58], [9], [46], [67], [33], [19], [38], [83].”
Hough Forests for Object Detection, Tracking, and Action Recognition (2011, background): “The resulting single detector is not only more convenient, but also achieves higher accuracy in multi-aspect detection tasks.”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, methodology): “Leibe et al. (2008b) focus on autonomous vehicle applications and propose a non-Markovian hypothesis selection framework”
Multimodal information fusion for urban scene understanding (2016, methodology): “Other methods use geometric cues like the ground plane to infer constraints for object detection [21].”
Simulation of pedestrian dynamics using a two dimensional cellular automaton (2001, unknown): “The central element describes the probability for the particle not to move at all, the remaining 8 correspond to a move to the neighbouring cells.”
Discrete-continuous optimization for multi-target tracking (2012, background): “A slightly different approach is presented in [17, 18], where a redundant set of putative trajectories is pre-computed, and the optimization takes place at the trajectory level by pruning to an optimal subset, formalized as a quadratic Boolean problem.”
The Cityscapes Dataset (2015, methodology): “Furthermore, the task of a semantic understanding of urban scenarios is addressed by several datasets, such as CamVid [1], Leuven [13], and Daimler Urban Segmentation [21].”