Learning to Optimize Non-Rigid Tracking

Li, Yang; Božič, Aljaž; Zhang, Tianwei; Ji, Yanli; Harada, Tatsuya; NieBner, Matthias

doi:10.1109/cvpr42600.2020.00496

Cited by 26 publications

(18 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 6 Tracking result of foreman sequences (A) using mean-shift tracking (B) using proposed algorithm (Frames 1,6,51,84,114,126 are displayed). Image credit: Li et al (2020).…”

Section: Resultsmentioning

confidence: 99%

“…They used visual information from the RGB-depth camera and evolved the contour by calculating the contour forces. Li et al (2020) generated pre-conditioner from a large number of training samples and proposed conditionNet in their process and increased the accuracy and speed of non-rigid RGB-depth object tracking process employing a learnable optimization process.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A fast region-based active contour for non-rigid object tracking and its shape retrieval

Mewada

Al‐Asad

Patel

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

Conventional tracking approaches track objects using a rectangle bounding box. Gait, gesture and many medical analyses require non-rigid shape extraction. A non-rigid object tracking is more difficult because it needs more accurate object shape and background separation in contrast to rigid bounding boxes. Active contour plays a vital role in the retrieval of image shape. However, the large computation time involved in contour tracing makes its use challenging in video processing. This paper proposes a new formation of the region-based active contour model (ACM) using a mean-shift tracker for video object tracking and its shape retrieval. The removal of re-initialization and fast deformation of the contour is proposed to retrieve the shape of the desired object. A contour model is further modified using a mean-shift tracker to track and retrieve shape simultaneously. The experimental results and their comparative analysis concludes that the proposed contour-based tracking succeed to track and retrieve the shape of the object with 71.86% accuracy. The contour-based mean-shift tracker resolves the scale-orientation selection problem in non-rigid object tracking, and resolves the weakness of the erroneous localization of the object in the frame by the tracker.

show abstract

“…Figure 6 Tracking result of foreman sequences (A) using mean-shift tracking (B) using proposed algorithm (Frames 1,6,51,84,114,126 are displayed). Image credit: Li et al (2020).…”

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A fast region-based active contour for non-rigid object tracking and its shape retrieval

Mewada

Al‐Asad

Patel

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Using deep learning, DeepDeform [5] replaces the classical feature matching by CNN-based correspondence matching. Li et al [30] goes one step further and differentiates through the N-ICP algorithm thus obtaining a dense feature matching term. A similar direction is taken by Neural Non-Rigid Tracking [4]; however, their focus lies in an end-to-end robust correspondence estimation.…”

Section: Non-rigid Tracking Using Depth Sensorsmentioning

confidence: 99%

4DComplete: Non-Rigid Motion Estimation Beyond the Observable Surface

Li¹,

Takehara²,

Taketomi³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…DeepDeform [10] learns sparse global correspondence for patches in non-rigid deforming RGB-D sequences. Li et al [37] learns non-rigid features through a differentiable non-rigid alignment optimization. NNRT [9] focuses on end-to-end robust correspondence estimation with an outlier rejection network.…”

Section: Non-rigid Correspondencementioning

confidence: 99%