Search Results for author: Mehdi S. M. Sajjadi

Found 23 papers, 9 papers with code

DyST: Towards Dynamic Neural Scene Representations on Real-World Videos

no code implementations9 Oct 2023 Maximilian Seitzer, Sjoerd van Steenkiste, Thomas Kipf, Klaus Greff, Mehdi S. M. Sajjadi

Our Dynamic Scene Transformer (DyST) model leverages recent work in neural scene representation to learn a latent decomposition of monocular real-world videos into scene content, per-view scene dynamics, and camera pose.

DORSal: Diffusion for Object-centric Representations of Scenes et al

no code implementations13 Jun 2023 Allan Jabri, Sjoerd van Steenkiste, Emiel Hoogeboom, Mehdi S. M. Sajjadi, Thomas Kipf

In this paper, we leverage recent progress in diffusion models to equip 3D scene representation learning models with the ability to render high-fidelity novel views, while retaining benefits such as object-level scene editing to a large degree.

Neural Rendering Representation Learning +2

Sensitivity of Slot-Based Object-Centric Models to their Number of Slots

no code implementations30 May 2023 Roland S. Zimmermann, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Thomas Kipf, Klaus Greff

Self-supervised methods for learning object-centric representations have recently been applied successfully to various datasets.

RePAST: Relative Pose Attention Scene Representation Transformer

no code implementations3 Apr 2023 Aleksandr Safin, Daniel Duckworth, Mehdi S. M. Sajjadi

The Scene Representation Transformer (SRT) is a recent method to render novel views at interactive rates.

Invariant Slot Attention: Object Discovery with Slot-Centric Reference Frames

1 code implementation9 Feb 2023 Ondrej Biza, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gamaleldin F. Elsayed, Aravindh Mahendran, Thomas Kipf

Automatically discovering composable abstractions from raw perceptual data is a long-standing challenge in machine learning.

Object Discovery

RUST: Latent Neural Scene Representations from Unposed Imagery

no code implementations CVPR 2023 Mehdi S. M. Sajjadi, Aravindh Mahendran, Thomas Kipf, Etienne Pot, Daniel Duckworth, Mario Lucic, Klaus Greff

Our main insight is that one can train a Pose Encoder that peeks at the target image and learns a latent pose embedding which is used by the decoder for view synthesis.

Novel View Synthesis

Object Scene Representation Transformer

no code implementations14 Jun 2022 Mehdi S. M. Sajjadi, Daniel Duckworth, Aravindh Mahendran, Sjoerd van Steenkiste, Filip Pavetić, Mario Lučić, Leonidas J. Guibas, Klaus Greff, Thomas Kipf

A compositional understanding of the world in terms of objects and their geometry in 3D space is considered a cornerstone of human cognition.

Novel View Synthesis Representation Learning

Test-time Adaptation with Slot-Centric Models

1 code implementation21 Mar 2022 Mihir Prabhudesai, Anirudh Goyal, Sujoy Paul, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gaurav Aggarwal, Thomas Kipf, Deepak Pathak, Katerina Fragkiadaki

In our work, we find evidence that these losses are insufficient for the task of scene decomposition, without also considering architectural inductive biases.

Image Classification Image Segmentation +7

RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs

no code implementations CVPR 2022 Michael Niemeyer, Jonathan T. Barron, Ben Mildenhall, Mehdi S. M. Sajjadi, Andreas Geiger, Noha Radwan

We observe that the majority of artifacts in sparse input scenarios are caused by errors in the estimated scene geometry, and by divergent behavior at the start of training.

Novel View Synthesis

Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representations

1 code implementation CVPR 2022 Mehdi S. M. Sajjadi, Henning Meyer, Etienne Pot, Urs Bergmann, Klaus Greff, Noha Radwan, Suhani Vora, Mario Lucic, Daniel Duckworth, Alexey Dosovitskiy, Jakob Uszkoreit, Thomas Funkhouser, Andrea Tagliasacchi

In this work, we propose the Scene Representation Transformer (SRT), a method which processes posed or unposed RGB images of a new area, infers a "set-latent scene representation", and synthesises novel views, all in a single feed-forward pass.

Novel View Synthesis Semantic Segmentation

NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections

1 code implementation CVPR 2021 Ricardo Martin-Brualla, Noha Radwan, Mehdi S. M. Sajjadi, Jonathan T. Barron, Alexey Dosovitskiy, Daniel Duckworth

We present a learning-based method for synthesizing novel views of complex scenes using only unstructured collections of in-the-wild photographs.

From Variational to Deterministic Autoencoders

4 code implementations ICLR 2020 Partha Ghosh, Mehdi S. M. Sajjadi, Antonio Vergari, Michael Black, Bernhard Schölkopf

Variational Autoencoders (VAEs) provide a theoretically-backed and popular framework for deep generative models.

Density Estimation

Perceptual Video Super Resolution with Enhanced Temporal Consistency

no code implementations20 Jul 2018 Eduardo Pérez-Pellitero, Mehdi S. M. Sajjadi, Michael Hirsch, Bernhard Schölkopf

Together with a video discriminator, we also propose additional loss functions to further reinforce temporal consistency in the generated sequences.

Image Super-Resolution Video Super-Resolution

Assessing Generative Models via Precision and Recall

4 code implementations NeurIPS 2018 Mehdi S. M. Sajjadi, Olivier Bachem, Mario Lucic, Olivier Bousquet, Sylvain Gelly

Recent advances in generative modeling have led to an increased interest in the study of statistical divergences as means of model comparison.

Tempered Adversarial Networks

no code implementations ICML 2018 Mehdi S. M. Sajjadi, Giambattista Parascandolo, Arash Mehrjou, Bernhard Schölkopf

A possible explanation for training instabilities is the inherent imbalance between the networks: While the discriminator is trained directly on both real and fake samples, the generator only has control over the fake samples it produces since the real data distribution is fixed by the choice of a given dataset.

Frame-Recurrent Video Super-Resolution

no code implementations CVPR 2018 Mehdi S. M. Sajjadi, Raviteja Vemulapalli, Matthew Brown

Recent advances in video super-resolution have shown that convolutional neural networks combined with motion compensation are able to merge information from multiple low-resolution (LR) frames to generate high-quality images.

Motion Compensation Multi-Frame Super-Resolution +1

Depth Estimation Through a Generative Model of Light Field Synthesis

no code implementations6 Sep 2016 Mehdi S. M. Sajjadi, Rolf Köhler, Bernhard Schölkopf, Michael Hirsch

Light field photography captures rich structural information that may facilitate a number of traditional image processing and computer vision tasks.

Depth Estimation

Peer Grading in a Course on Algorithms and Data Structures: Machine Learning Algorithms do not Improve over Simple Baselines

no code implementations2 Jun 2015 Mehdi S. M. Sajjadi, Morteza Alamgir, Ulrike Von Luxburg

Peer grading is the process of students reviewing each others' work, such as homework submissions, and has lately become a popular mechanism used in massive open online courses (MOOCs).

Cannot find the paper you are looking for? You can Submit a new open access paper.