Kernel Cross-Correlator

By Chen Wang, Le Zhang, Lihua Xie, Junsong Yuan. 2017

109 citations recorded by the source corpus.

Cross-correlator plays a significant role in many visual perception tasks, such as object detection and tracking. Beyond the linear cross-correlator, this paper proposes a kernel cross-correlator (KCC) that breaks traditional limitations. First, by introducing the kernel trick, the KCC extends the linear cross-correlation to non-linear space, which is more robust to signal noises and distortions. Second, the connection to the existing works shows that KCC provides a unified solution for correlation filters. Third, KCC is applicable to any kernel function and is not limited to circulant structure on training data, thus it is able to predict affine transformations with customized properties. Last, by leveraging the fast Fourier transform (FFT), KCC eliminates direct calculation of kernel vectors, thus achieves better performance yet still with a reasonable computational cost. Comprehensive experiments on visual tracking and human activity recognition using wearable devices demonstrate its robustness, flexibility, and efficiency. The source codes of both experiments are released at https://github.com/wang-chen/KCC.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

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

1. Junsong Yuan
2. Lihua Xie
3. Chen Wang
4. Yuheng Qiu
5. Sebastian A. Scherer
6. Lei Tai
7. Yin Zhou
8. Oncel Tuzel

In Computer Vision and Pattern Recognition

1. Sebastian A. Scherer
2. Yuheng Qiu
3. Chen Wang
4. Junsong Yuan
5. Yafei Hu
6. Wenshan Wang
7. Lihua Xie
8. Jianfei Yang

Citations in the corpus

• Lifelong Graph Learning (2020, neutral): “We take every 25 sequential data points from a single sensor as a node and perform recognition for every 50 sequential points (1.67s) shown in Figure 2b, which is also adopted by [44].”
• Visual Memorability for Robotic Interestingness via Unsupervised Online Learning (2020, methodology): “Recall that 2D cross-correlation is the inner-products between the first signal and circular translations of the second signal [38], we can compute Si(x,Mi) as”
• Kervolutional Neural Networks (2019, background): “Interestingly, the kernel crosscorrelator (KCC) defined in [54] is equivariant to any affine transforms (e.”
• AutoTrack: Towards High-Performance Visual Tracking for UAV With Automatic Spatio-Temporal Regularization (2020, unknown): “Twelve real-time trackers (with a speed of >30fps), i.e., KCF [7], DCF [7], KCC [57] fDSST [11], DSST [58], BACF [22], STAPLE-CA [10], STAPLE [10], MCCTH [13], STRCF [14], ECO-HC [20], ARCF-H [26], and five non-real-time ones, i.e., SRDCF [8], SAMF [6], CSRDCF [30], SRDCFdecon [24], ARCF-HC [26] are”