Search Results for author:
Found 8 papers, 5 papers with code
GAIA: Zero-shot Talking Avatar Generation
Zero-shot talking avatar generation aims at synthesizing natural talking videos from speech and a single portrait image.
Unlimited-Size Diffusion Restoration
Our simple, parameter-free approaches can be used not only for image restoration but also for image generation of unlimited sizes, with the potential to be a general tool for diffusion models.
LaPE: Layer-adaptive Position Embedding for Vision Transformers with Independent Layer Normalization
A typical way to introduce position information is adding the absolute Position Embedding (PE) to patch embedding before entering VTs.
Position Embedding Needs an Independent Layer Normalization
By analyzing the input and output of each encoder layer in VTs using reparameterization and visualization, we find that the default PE joining method (simply adding the PE and patch embedding together) operates the same affine transformation to token embedding and PE, which limits the expressiveness of PE and hence constrains the performance of VTs.
ACSeg: Adaptive Conceptualization for Unsupervised Semantic Segmentation
Recently, self-supervised large-scale visual pre-training models have shown great promise in representing pixel-level semantic relationships, significantly promoting the development of unsupervised dense prediction tasks, e. g., unsupervised semantic segmentation (USS).
Locality Guidance for Improving Vision Transformers on Tiny Datasets
Therefore, our locality guidance approach is very simple and efficient, and can serve as a basic performance enhancement method for VTs on tiny datasets.
$L_2$BN: Enhancing Batch Normalization by Equalizing the $L_2$ Norms of Features
In this paper, we analyze batch normalization from the perspective of discriminability and find the disadvantages ignored by previous studies: the difference in $l_2$ norms of sample features can hinder batch normalization from obtaining more distinguished inter-class features and more compact intra-class features.
On the Global Optima of Kernelized Adversarial Representation Learning
Numerical experiments on UCI, Extended Yale B and CIFAR-100 datasets indicate that, (a) practically, our solution is ideal for "imparting" provable invariance to any biased pre-trained data representation, and (b) empirically, the trade-off between utility and invariance provided by our solution is comparable to iterative minimax optimization of existing deep neural network based approaches.
