Fusing points and lines for high performance tracking

By Edward Rosten, Tom Drummond. Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1, 2005

1,332 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/bd63fcd5ec4ead649a32d62d005c9f3a410f4f31

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

Automatic Classification of Defective Photovoltaic Module Cells in Electroluminescence Images (2018, positive): “a random forest classifier trained on a set of corner features that is known as Features from Accelerated Segment Test (FAST) [23, 24].”
Particle Filter SLAM with High Dimensional Vehicle Model (2009, background): “There are more elaborate detectors available, for example, the scale-invariant feature transform (sift) [24], the speeded up robust features (surf) [4], or the fast corner detector [28, 29].”
Keypoint recognition using randomized trees (2006, methodology): “Following a philosophy similar to the one proposed in [ 26 ], for each pixel m, we consider the”
Scale Drift-Aware Large Scale Monocular SLAM (2010, background): “In order to speed-up the search, we do not apply the matching at every single pixel in the search region, but only there where FAST features [26] are detected.”
Efficient Optical Flow and Stereo Vision for Velocity Estimation and Obstacle Avoidance on an Autonomous Pocket Drone (2016, background): “Since tailed flapping wing MAVs such as the DelFly Explorer are passively stable, there is no need to compute their velocity to compensate for drift.”
Recent progress in semantic image segmentation (2018, background): “…(Shi et al. 1994), Sub-pixelCorner (Medioni andYasumoto 1987), SUSAN (Smith and Brady 1997), Features from Accelerated Segment Test (FAST) (Rosten and Drummond 2005), FAST- ER (Rosten et al. 2010), AGAST (Mair et al. 2010) and Multiscale AGAST (Leutenegger et al. 2011) Detector,…”
Faster and Better: A Machine Learning Approach to Corner Detection (2008, background): “The original detector [95], [96] classifies p as a corner if there exists a set of n contiguous pixels in the circle which are all brighter than the intensity of the candidate pixel Ip plus a thresholdt, or all darker thanIp − t, as illustrated in Figure 1.”
R2D2: Repeatable and Reliable Detector and Descriptor (2019, methodology): “The ﬁrst approach for keypoint detection to rely on machine learning was FAST [41].”
Machine Learning for High-Speed Corner Detection (2006, background): “In addition, the approach allows the detector to be generalised, producing a suite of high-speed detectors which we currently use for real-time tracking [2] and AR label placement [3].”
Adaptive and Generic Corner Detection Based on the Accelerated Segment Test (2010, methodology): “With the introduction of recent techniques such as the “Features from Accelerated Segment Test” (FAST) [8], feature extraction has seen significant performance increase for real-time Computer Vision applications.”
Robust 3D visual tracking using particle filtering on the special Euclidean group: A combined approach of keypoint and edge features (2011, methodology): “To improve pose priors, there have been attempts to enhance the pose accuracy between frames by incorporating interest points [4], [5], [6] or employing additional sensors [7].”
Real-time monocular SLAM: Why filter? (2010, methodology): “However, each 3D point was only visible on average in three to four keyframes due to various reasons such as the restricted field of view, occlusion and the sensitivity of SURF and FAST to perspective transformations.”
Evaluation of Interest Point Detectors and Feature Descriptors for Visual Tracking (2011, background): “Rosten and Drummond (2005, 2006) developed this idea further and sped up the process by reducing the number of pixel tests with machine learning techniques.”
Local histogram matching for efficient optical flow computation applied to velocity estimation on pocket drones (2016, methodology): “In robotics, computational efficiency is important and hence sparse optical flow is often determined with the help of a feature detector such as Shi-Tomasi [5] or FAST [6], followed by Lucas-Kanade feature tracking [7].”