SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos

Zeng, Ailing; Yang, Lei; Ju, Xuan; Jiefeng, Li,; Wang, Jianyi; Xu, Qiang

doi:10.1007/978-3-031-20065-6_36

Cited by 36 publications

(7 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparison of the acceleration errors for our method and smoothnet. 24 Our method shows significantly lower acceleration errors than the SOTA method.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 77%

“…We compare the results with state-of-the-art frame-based human mesh recovery methods, [4][5][6][7][8][21][22][23][24] demonstrating the ability of our model to recover accurate and smooth 3D human motion from video. As shown in Table 1, our method shows good performance in the indoor dataset Human3.6M 18 and the challenging field datasets 3DPW 16 and MPI-INF 3DHP.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 97%

See 1 more Smart Citation

Temporally consistent 3D human motion estimation from a video

zhang,

Liu,

Wang

et al. 2024

Seventh International Conference on Computer Graphics and Virtuality (ICCGV 2024)

View full text Add to dashboard Cite

“…Comparison of the acceleration errors for our method and smoothnet. 24 Our method shows significantly lower acceleration errors than the SOTA method.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 77%

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 97%

Temporally consistent 3D human motion estimation from a video

zhang,

Liu,

Wang

et al. 2024

Seventh International Conference on Computer Graphics and Virtuality (ICCGV 2024)

View full text Add to dashboard Cite

“…L θ , L β , L cam represents L1 loss between the estimated human pose, shape and camera pose to our synthetic ground truth. L smooth is adopted from [65] by penalizing the velocity and acceleration between the estimation and the ground truth. For the neural descent module, the objective loss can be written as:…”

Section: Loss Functionmentioning

confidence: 99%

DeepFilter: A Deep Learning Based Variant Filter for VarDict

Zhang

Yin

Wei

et al. 2023

Tsinghua Sci. Technol.

View full text Add to dashboard Cite

With the development of sequencing technologies, somatic mutation analysis has become an important component in cancer research and treatment. VarDict is a commonly used somatic variant caller for this task.Although the heuristic-based VarDict algorithm exhibits high sensitivity and versatility, it may detect higher amounts of false positive variants than callers, limiting its clinical practicality. To address this problem, we propose DeepFilter, a deep-learning based filter for VarDict, which can filter out the false positive variants detected by VarDict effectively.Our approach trains two models for insertion-deletion mutations (InDels) and single nucleotide variants (SNVs), respectively. Experiments show that DeepFilter can filter at least 98.5% of false positive variants and retain 93.5% of true positive variants for InDels and SNVs in the commonly used tumor-normal paired mode. Source code and pre-trained models are available at https://github.com/LeiHaoa/DeepFilter.

show abstract

“…We follow MSCOCO [32] to define the first 17 keypoints and add 4 additional keypoints, left/right fingers and left/right toes, which is beneficial for 3D pose estimation and shape recovery by providing more comprehensive constraints [50,82]. Self-contact keypoints [4,45] also demonstrate benefits for 3D pose and shape estimation by disambiguating the body part depth unknown in 2D human pose representation, avoiding self-collisions and penetrations, and being easier to use than ordinal depth [51,59,88].…”

Section: Data Annotationmentioning

confidence: 99%

Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes

Ju¹,

Zeng²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

Humans have long been recorded in a variety of forms since antiquity. For example, sculptures and paintings were the primary media for depicting human beings before the invention of cameras. However, most current human-centric computer vision tasks like human pose estimation and human image generation focus exclusively on natural images in the real world. Artificial humans, such as those in sculptures, paintings, and cartoons, are commonly neglected, making existing models fail in these scenarios.As an abstraction of life, art incorporates humans in both natural and artificial scenes. We take advantage of it and introduce the Human-Art dataset to bridge related tasks in natural and artificial scenarios. Specifically, Human-Art contains 50k high-quality images with over 123k person instances from 5 natural and 15 artificial scenarios, which are annotated with bounding boxes, keypoints, self-contact points, and text information for humans represented in both 2D and 3D. It is, therefore, comprehensive and versatile for various downstream tasks. We also provide a rich set of baseline results and detailed analyses for related tasks, including human detection, 2D and 3D human pose estimation, image generation, and motion transfer. As a challenging dataset, we hope Human-Art can provide insights for relevant research and open up new research questions.

show abstract

SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos

Cited by 36 publications

References 48 publications

Temporally consistent 3D human motion estimation from a video

Temporally consistent 3D human motion estimation from a video

DeepFilter: A Deep Learning Based Variant Filter for VarDict

Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes

Contact Info

Product

Resources

About