Zobrazeno 1 - 10
of 50
pro vyhledávání: '"Balle, Johannes"'
We introduce a lightweight, flexible and end-to-end trainable probability density model parameterized by a constrained Fourier basis. We assess its performance at approximating a range of multi-modal 1D densities, which are generally difficult to fit
Externí odkaz:
http://arxiv.org/abs/2402.15345
We consider lossy compression of an information source when the decoder has lossless access to a correlated one. This setup, also known as the Wyner-Ziv problem, is a special case of distributed source coding. To this day, practical approaches for th
Externí odkaz:
http://arxiv.org/abs/2310.16961
We show how perceptual embeddings of the visual system can be constructed at inference-time with no training data or deep neural network features. Our perceptual embeddings are solutions to a weighted least squares (WLS) problem, defined at the pixel
Externí odkaz:
http://arxiv.org/abs/2310.05986
We introduce a distortion measure for images, Wasserstein distortion, that simultaneously generalizes pixel-level fidelity on the one hand and realism or perceptual quality on the other. We show how Wasserstein distortion reduces to a pure fidelity c
Externí odkaz:
http://arxiv.org/abs/2310.03629
We consider lossy compression of an information source when the decoder has lossless access to a correlated one. This setup, also known as the Wyner-Ziv problem, is a special case of distributed source coding. To this day, real-world applications of
Externí odkaz:
http://arxiv.org/abs/2305.04380
Artificial Neural-Network-based (ANN-based) lossy compressors have recently obtained striking results on several sources. Their success may be ascribed to an ability to identify the structure of low-dimensional manifolds in high-dimensional ambient s
Externí odkaz:
http://arxiv.org/abs/2205.08518
A significant bottleneck in federated learning (FL) is the network communication cost of sending model updates from client devices to the central server. We present a comprehensive empirical study of the statistics of model updates in FL, as well as
Externí odkaz:
http://arxiv.org/abs/2201.02664
Compressing the output of \epsilon-locally differentially private (LDP) randomizers naively leads to suboptimal utility. In this work, we demonstrate the benefits of using schemes that jointly compress and privatize the data using shared randomness.
Externí odkaz:
http://arxiv.org/abs/2111.00092
Autor:
Mentzer, Fabian, Agustsson, Eirikur, Ballé, Johannes, Minnen, David, Johnston, Nick, Toderici, George
We present the first neural video compression method based on generative adversarial networks (GANs). Our approach significantly outperforms previous neural and non-neural video compression methods in a user study, setting a new state-of-the-art in v
Externí odkaz:
http://arxiv.org/abs/2107.12038
It has been demonstrated many times that the behavior of the human visual system is connected to the statistics of natural images. Since machine learning relies on the statistics of training data as well, the above connection has interesting implicat
Externí odkaz:
http://arxiv.org/abs/2106.04427