The prime challenge in unsupervised domain adaptation (DA) is to mitigate the domain shift between the source and target domains.
Conventional domain adaptation (DA) techniques aim to improve domain transferability by learning domain-invariant representations; while concurrently preserving the task-discriminability knowledge gathered from the labeled source data.
To this end, we cast 3D pose learning as a self-supervised adaptation problem that aims to transfer the task knowledge from a labeled source domain to a completely unpaired target.
Ranked #4 on Unsupervised 3D Human Pose Estimation on Human3.6M
Articulation-centric 2D/3D pose supervision forms the core training objective in most existing 3D human pose estimation techniques.
The advances in monocular 3D human pose estimation are dominated by supervised techniques that require large-scale 2D/3D pose annotations.
Ranked #7 on Unsupervised 3D Human Pose Estimation on Human3.6M
However, we find that latent features derived from the Fourier-based amplitude spectrum of deep CNN features hold a more tractable mapping with domain discrimination.
Unsupervised domain adaptation (DA) has gained substantial interest in semantic segmentation.
Ranked #1 on Domain Generalization on GTA5-to-Cityscapes
Thus, we aim to utilize implicit alignment without additional training objectives to perform adaptation.
We present a self-supervised human mesh recovery framework to infer human pose and shape from monocular images in the absence of any paired supervision.
Our approach not only generalizes to in-the-wild images, but also yields a superior trade-off between speed and performance, compared to prior top-down approaches.
We introduce a practical Domain Adaptation (DA) paradigm called Class-Incremental Domain Adaptation (CIDA).
However, generalizability of human pose estimation models developed using supervision on large-scale in-studio datasets remains questionable, as these models often perform unsatisfactorily on unseen in-the-wild environments.
Modeling dynamics of human motion is one of the most challenging sequence modeling problem, with diverse applications in animation industry, human-robot interaction, motion-based surveillance, etc.
Addressing this, we introduce a practical DA paradigm where a source-trained model is used to facilitate adaptation in the absence of the source dataset in future.
1) In the Procurement stage, we aim to equip the model for future source-free deployment, assuming no prior knowledge of the upcoming category-gap and domain-shift.
Camera captured human pose is an outcome of several sources of variation.
In this paper, we propose UM-Adapt - a unified framework to effectively perform unsupervised domain adaptation for spatially-structured prediction tasks, simultaneously maintaining a balanced performance across individual tasks in a multi-task setting.
Despite the remarkable success of generative adversarial networks, their performance seems less impressive for diverse training sets, requiring learning of discontinuous mapping functions.
In this work we propose a novel temporal pose-sequence modeling framework, which can embed the dynamics of 3D human-skeleton joints to a continuous latent space in an efficient manner.
The discriminator is trained also to regress this extrinsic factor r, which is used alongside with the intrinsic factor (encoded starting pose sequence) to generate a particular pose sequence.
In this work, we propose a data-efficient method which utilizes the geometric regularity of intraclass objects for pose estimation.
Such image comparison based approach also alleviates the problem of data scarcity and hence enhances scalability of the proposed approach for novel object categories with minimal annotation.
Supervised deep learning methods have shown promising results for the task of monocular depth estimation; but acquiring ground truth is costly, and prone to noise as well as inaccuracies.