A new framework for on-line object tracking based on SURF

Miao, Qiguang; Wang, Guijin; Shi, Chenbo; Lin, Xiao; Ruan, Zhiwei

doi:10.1016/j.patrec.2011.05.017

Cited by 37 publications

(20 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The non-rigid object trackers cited in [1], [2] are not appropriate for comparison. Thus we use Grabner's tracker [3] and our original tracker [6]. The speed of Grabner's tracker is 15fps on a PC with 2.93 GHz CPU and 4 GB RAM.…”

Section: Resultsmentioning

confidence: 99%

“…The proposed method is classified as a rigid object tracker. In contrast to the previous methods [5], [6], the novelty and contributions of this paper are twofold. First, we propose a new discriminative feature modeling and classification mechanism instead of the previously used boosting [4].…”

Section: Introductionmentioning

confidence: 99%

“…Thus the corresponding tracker cannot handle complex variations. Afterwards Miao et al [5], [6] employed the scale and rotation information of SURF features [7] to guide the on-line classifier learning process. Although the resulting tracking is robust to significant changes like scale, rotation and viewpoint, it cannot ensure long-term accurate tracking.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On-Line Rigid Object Tracking via Discriminative Feature Classification

Miao

Shi

Meng³

et al. 2016

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYThis paper proposes an on-line rigid object tracking framework via discriminative object appearance modeling and learning. Strong classifiers are combined with 2D scale-rotation invariant local features to treat tracking as a keypoint matching problem. For on-line boosting, we correspond a Gaussian mixture model (GMM) to each weak classifier and propose a GMM-based classifying mechanism. Meanwhile, selforganizing theory is applied to perform automatic clustering for sequential updating. Benefiting from the invariance of the SURF feature and the proposed on-line classifying technique, we can easily find reliable matching pairs and thus perform accurate and stable tracking. Experiments show that the proposed method achieves better performance than previously reported trackers.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On-Line Rigid Object Tracking via Discriminative Feature Classification

Miao

Shi

Meng³

et al. 2016

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The object region is transformed using the homography estimated based on the matching candidates. In [7], we further expand the idea and provide in detail a new framework.…”

Section: On-line Boosting For Object Trackingmentioning

confidence: 99%

“…Furthermore, the responses are distributed in horizontal and vertical direction in relation to the dominant orientation. Experiments show the resulted tracker is able to handle viewpoint change, partial occlusion and complex transformations [7].…”

Section: C(x)mentioning

confidence: 99%

Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA

Miao

Wang

Lin

2011

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYObject tracking is a major technique in image processing and computer vision. Tracking speed will directly determine the quality of applications. This paper presents a parallel implementation for a recently proposed scale-and rotation-invariant on-line object tracking system. The algorithm is based on NVIDIA's Graphics Processing Units (GPU) using Compute Unified Device Architecture (CUDA), following the model of single instruction multiple threads. Specifically, we analyze the original algorithm and propose the GPU-based parallel design. Emphasis is placed on exploiting the data parallelism and memory usage. In addition, we apply optimization technique to maximize the utilization of NVIDIA's GPU and reduce the data transfer time. Experimental results show that our GPGPUbased method running on a GTX480 graphics card could achieve up to 12X speed-up compared with the efficiency equivalence on an Intel E8400 3.0 GHz CPU, including I/O time.

show abstract

A double circle structure descriptor and Hough voting matching for real-time object detection

Liu

2016

Pattern Anal Applic

View full text Add to dashboard Cite

A new framework for on-line object tracking based on SURF

Cited by 37 publications

References 24 publications

On-Line Rigid Object Tracking via Discriminative Feature Classification

On-Line Rigid Object Tracking via Discriminative Feature Classification

Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA

A double circle structure descriptor and Hough voting matching for real-time object detection

Contact Info

Product

Resources

About