Human Detection Based on a Probabilistic Assembly of Robust Part Detectors

By Krystian Mikolajczyk, Cordelia Schmid, Andrew Zisserman. European Conference on Computer Vision, 2004

777 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f9cc6155cd69023a736a7b8f8680bcd6232c840e

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Detection and Tracking of Multiple, Partially Occluded Humans by Bayesian Combination of Edgelet based Part Detectors (2007, neutral): “If inter-object occlusion is present, the assumption of conditional independence between individual human appearances given the state, as in Mikolajczyk et al. (2004) , is not valid and a more complex formulation is necessary.”
A Survey of Recent Advances in Face Detection (2010, background): “Part-based Wavelet localized parts [83, 82] approaches SVM component detectors adaptively trained [28] Overlapping part detectors [61]”
FDDB: A benchmark for face detection in unconstrained settings (2010, background): “• Mikolajczyk’s face detector [14] 5 – we set the parameter for the minimum distance between eyes in a detected face to 5 pixels.”
Detection of multiple, partially occluded humans in a single image by Bayesian combination of edgelet part detectors (2005, methodology): “Nonetheless, we show a comparison with methods in [1] and [4] as they both use the MIT dataset that was available to us.”
Recent advances and trends in visual tracking: A review (2011, background): “Mikolajczyk et al. [27] introduced position-orientation histogram features for human detection.”
Robust Fragments-based Tracking using the Integral Histogram (2006, neutral): “We present extensive experimental results on challenging sequences, which demonstrate the robust tracking achieved by our algorithm (even with the use of only gray-scale (noncolor) information).”
A discriminative deep model for pedestrian detection with occlusion handling (2012, methodology): “Recent deformable models for object detection mainly model the translational deformation of parts [12, 21, 34].”
LIFT: Learned Invariant Feature Transform (2016, methodology): “We use a maximum of 1000 keypoints per image, and follow the standard evaluation protocol of [32] on the common viewpoint region.”
A survey on face detection in the wild: Past, present and future (2015, methodology): “Other notable parts-based methods include [49] [50] [51].”
Joint Cascade Face Detection and Alignment (2014, result): “On FDDB dataset, we compare with all the published results in the platform [26,32,24,14,23,12,16,8,25,18,30,13,15].”
Strike a pose: tracking people by finding stylized poses (2005, background): “In general, detecting people is hard because people wear different clothes and can take on many poses; this suggests a bottom-up approach using part detectors [9, 11, 13, 17].”
Deep Learning Strong Parts for Pedestrian Detection (2015, methodology): “Part-Based Pedestrian Detectors One stream of partbased approaches [20, 19, 9, 37] firstly trained part detectors in a fully supervised manner and then combined their outputs to fit a geometric model.”
Human detection from images and videos: A survey (2016, methodology): “Its performance was also verified in the work of Mikolajczyk and Schmid [101] and used to encode the parts of the object in [102, 140].”
Geometric context from a single image (2005, background): “This proves a severe limitation, preventing important information – scale relationships, surface orientations, free-space reasoning, etc. – from ever being captured.”