Lower Dimensional Kernels for Video Discriminators

By Emmanuel Kahembwe, Subramanian Ramamoorthy. 2019

53 citations recorded by the source corpus.

This work presents an analysis of the discriminators used in Generative Adversarial Networks (GANs) for Video. We show that unconstrained video discriminator architectures induce a loss surface with high curvature which make optimization difficult. We also show that this curvature becomes more extreme as the maximal kernel dimension of video discriminators increases. With these observations in hand, we propose a methodology for the design of a family of efficient Lower-Dimensional Video Discriminators for GANs (LDVD-GANs). The proposed methodology improves the performance and efficiency of video GAN models it is applied to and demonstrates good performance on complex and diverse datasets such as UCF-101. In particular, we show that LDVDs can double the performance of Temporal-GANs and provide for state-of-the-art performance on a single GPU using the proposed methodology.

Subfields: Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Generative Models

Citations in the corpus

Video Diffusion Models (2022, neutral): “The Inception Score we calculate for real data ( ≈ 60 ) is consistent with that reported by [26], who also report a higher real data Inception score of ≈ 90 for data sampled at the 128x128 resolution, which indicates that our 64x64 model might be at a disadvantage compared to works that generate at…”
StyleGAN-V: A Continuous Video Generator with the Price, Image Quality and Perks of StyleGAN2 (2021, unknown): “3 [26] Emmanuel Kahembwe and Subramanian Ramamoorthy.”
Align Your Latents: High-Resolution Video Synthesis with Latent Diffusion Models (2023, unknown): “[37] Emmanuel Kahembwe and Subramanian Ramamoorthy.”
Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models (2023, unknown): “18 - LDVD-GAN [20] 22.91 ± .”
Latte: Latent Diffusion Transformer for Video Generation (2024, unknown): “Firstly, several studies have sought to extend the capabilities of powerful GAN-based image generators to create videos (Vondrick et al, 2016; Saito et al, 2017; Wang et al, 2020b,a; Kahembwe and Ramamoorthy, 2020).”
Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer (2022, background): “28, which is close to the upper bound shown in the “Real data” row [2,27].”