Object Detection with Discriminatively Trained Part-Based Models

Felzenszwalb, Pedro F.; Girshick, Ross; McAllester, David; Ramanan, Deva

doi:10.1109/tpami.2009.167

Cited by 8,868 publications

(5,589 citation statements)

References 41 publications

Supporting

Mentioning

5,366

Contrasting

Unclassified

Order By: Relevance

“…Kwon et al [39] represented an object by a fixed number of local patches and updated the model during tracking. In contrast, Felzenszwalb et al [48] developed a multi-scale deformable part model to detect and localize objects of a generic category. The part model is composed of a coarse global template for the entire object and high-resolution part templates.…”

Section: Fragment-based Trackingmentioning

confidence: 99%

“…For most of the recently reported algorithms [48][49][50][51][52][53][54][55][56][57][58][59]58], in contrast, spatial constraints between these fragments are often imposed, and these algorithms are more robust to deformation and illumination changes. Among these algorithms, graph representation-based method is a particularly popular method, which include inter-ARG and intra-ARG [51], attributed relational feature graph (ARFG) [52], dynamic graph(DG) [56], memory graph [57], star model [48], etc. Although graph representation is efficient compared with pixel-based representation, since it only requires a small number of features to model an object, these methods rely on the reliability of local features.…”

Section: Fragment-based Trackingmentioning

confidence: 99%

See 1 more Smart Citation

Structured fragment-based object tracking using discrimination, uniqueness, and validity selection

Zheng

Xin

et al. 2017

Multimedia Systems

View full text Add to dashboard Cite

Local features have widely been used in visual tracking to improve robustness in the presence of partial occlusion, deformation, and rotation. In this paper, a local fragment-based object tracking algorithm is proposed. Unlike many existing fragment-based algorithms using all the fragments and allocating the weight to each fragment according to similarity, the proposed algorithm only selects discriminative, unique, and valid fragments for tracking. First, discrimination and uniqueness metric are defined for each local fragment, and an automatic pre-selection mechanism is proposed for all these fragments. Second, a Harris-SIFT filter is used to select the current valid fragments and exclude the occluded or highly deformed fragments. By selecting the discriminative, unique, and valid fragments, these fragments are used to construct a structured description for the object. Finally, the object tracking is performed using the selected fragments combining the displacement and similarity, as well as spatial constraint of the selected fragments. The object template can be updated by fusing feature similarity and structural consistency. The experimental results on a recent OTB 2013 tracking benchmark data set demonstrate that the proposed algorithm can achieve reliable tracking results even in the presence of significant appearance changes, partial occlusion, and similar disturbances.

show abstract

Section: Fragment-based Trackingmentioning

confidence: 99%

Section: Fragment-based Trackingmentioning

confidence: 99%

Structured fragment-based object tracking using discrimination, uniqueness, and validity selection

Zheng

Xin

et al. 2017

Multimedia Systems

View full text Add to dashboard Cite

show abstract

“…Danelljan et al [26] employed an adaptive feature dimensionality reduction method as in Ref. [27] to reduce the computational cost, while tracking the performance is preserved. A collaborative correlation tracker is proposed in [28].…”

Section: Scale Estimation In Dcf-based Visual Trackingmentioning

confidence: 99%

A comparison study on the adaptive scale estimation of correlation filter-based visual tracking methods

Wang¹,

Cai²,

Wang³

et al. 2017

2017 IEEE International Conference on Real-Time Computing and Robotics (RCAR)

View full text Add to dashboard Cite

Recently, discriminative correlation filter-based method becomes one of the popular directions in the field of visual tracking because of its computational efficiency and excellent performance, which make it especially suitable for realtime application. Most of them are focused only on the transition estimation. However, accurate scale estimation of the target plays a very important role in long-term tracking task and is still a challenging problem. The principle of CFbased visual tracking is introduced first. The approaches of adaptive scale estimation in correlation filter-based visual tracking methods are summarized in this paper, and their performances are analyzed by experiment comparison. The works here can provide a better understanding on the scale estimation problem for correlation filter-based visual tracking. Furthermore, maybe with the same strategy, other factors in visual tracking, such as appearance variation, can be integrated into the framework to improve the performance of correlation filter-based method.

show abstract

“…A significant amount of research has been carried out to mitigate partial occlusion by detecting only part of the body such as heads, faces, eyes and head-shoulders [18]- [20]. The shape of people's heads changes or differ with hair styles and head coverings.…”

Section: Part Body Detection Based Algorithmsmentioning

confidence: 99%