Search Results for author:
Found 10 papers, 1 papers with code
AdamL: A fast adaptive gradient method incorporating loss function
Adaptive first-order optimizers are fundamental tools in deep learning, although they may suffer from poor generalization due to the nonuniform gradient scaling.
ReCLIP: Refine Contrastive Language Image Pre-Training with Source Free Domain Adaptation
Large-scale Pre-Training Vision-Language Model such as CLIP has demonstrated outstanding performance in zero-shot classification, e. g. achieving 76. 3% top-1 accuracy on ImageNet without seeing any example, which leads to potential benefits to many tasks that have no labeled data.
On the Convergence of the Gradient Descent Method with Stochastic Fixed-point Rounding Errors under the Polyak-Lojasiewicz Inequality
When training neural networks with low-precision computation, rounding errors often cause stagnation or are detrimental to the convergence of the optimizers; in this paper we study the influence of rounding errors on the convergence of the gradient descent method for problems satisfying the Polyak-Lojasiewicz inequality.
CameraPose: Weakly-Supervised Monocular 3D Human Pose Estimation by Leveraging In-the-wild 2D Annotations
By adding a camera parameter branch, any in-the-wild 2D annotations can be fed into our pipeline to boost the training diversity and the 3D poses can be implicitly learned by reprojecting back to 2D.
Ranked #68 on
3D Human Pose Estimation
on Human3.6M
On the influence of stochastic roundoff errors and their bias on the convergence of the gradient descent method with low-precision floating-point computation
When implementing the gradient descent method in low precision, the employment of stochastic rounding schemes helps to prevent stagnation of convergence caused by the vanishing gradient effect.
A Simple and Efficient Stochastic Rounding Method for Training Neural Networks in Low Precision
Conventional stochastic rounding (CSR) is widely employed in the training of neural networks (NNs), showing promising training results even in low-precision computations.
Improved stochastic rounding
When a sequence of computations is implemented, round-off errors may be magnified or accumulated.
Early Recognition of Human Activities from First-Person Videos Using Onset Representations
In this paper, we propose a methodology for early recognition of human activities from videos taken with a first-person viewpoint.
Spatio-temporal Depth Cuboid Similarity Feature for Activity Recognition Using Depth Camera
Local spatio-temporal interest points (STIPs) and the resulting features from RGB videos have been proven successful at activity recognition that can handle cluttered backgrounds and partial occlusions.
View invariant human action recognition using histograms of 3D joints
Our method is real-time and achieves superior results on the challenging 3D action dataset.
Ranked #3 on
Skeleton Based Action Recognition
on UWA3D
