Maneesh Kumar Singh

University of Illinois at Urbana-Champaign, Ece, Alumnus

University of Illinois at Urbana-Champaign, Electrical and Computer Engineering, Alumnus

Followers

Following

Public Views

He has more than 15 years of R&D experience in designing, testing and performance analysis of statistical computer vision systems. He has more than 20 peer-reviewed publications as well 3 patents in following focus areas.

Focus areas: nonparametric generative models and statistics with applications in real-time vision for surveillance and security, industrial inspection and medical diagnostics.

Research and technology development experience:

- Low-level Vision designing features and features detectors; image segmentation and clustering; image analysis, restoration and denoising.

- Nonparametric and Semiparametric methods Data modeling using kernels, kernel density estimation, robust estimation, order statistics, classification.

- Nonlinear optimization and Variational Methods Non-convex optimization using
bounds; level sets; Mean Shift.

- Image and Video Analytics – Object detection and recognition, tracking, change detection, image registration and 3D from stereo.
Supervisors: Prof. Dr. Narendra Ahuja

less

Kwonjoon Lee

Jonas Wildberger

Andrea Asperti

Università di Bologna

Université du Luxembourg

Xinqi Zhu

The University of Sydney

InterestsView All (11)

Uploads

Papers by Maneesh Kumar Singh

Wavelets to the Rescue: Improving Sample Quality of Latent Variable Deep Generative Models

Variational Autoencoders (VAE) are probabilistic deep generative models underpinned by elegant th... more Variational Autoencoders (VAE) are probabilistic deep generative models underpinned by elegant theory, stable training processes, and meaningful manifold representations. However, they produce blurry images due to a lack of explicit emphasis over high-frequency textural details of the images, and the difficulty to directly model the complex joint probability distribution over the high-dimensional image space. In this work, we approach these two challenges with a novel wavelet space VAE that uses the decoder to model the images in the wavelet coefficient space. This enables the VAE to emphasize over high-frequency components within an image obtained via wavelet decomposition. Additionally, by decomposing the complex function of generating high-dimensional images into inverse wavelet transformation and generation of wavelet coefficients, the latter becomes simpler to model by the VAE. We empirically validate that deep generative models operating in the wavelet space can generate images of higher quality than the image (RGB) space counterparts. Quantitatively, on benchmark natural image datasets, we achieve consistently better FID scores than VAE based architectures and competitive FID scores with a variety of GAN models for the same architectural and experimental setup. Furthermore, the proposed wavelet-based generative model retains desirable attributes like disentangled and informative latent representation without losing the quality in the generated samples.

Maneesh Kumar Singh

Uploads

Papers by Maneesh Kumar Singh

Log In