64 citations · 70 across the 8 of their papers we have counts for
9 papers
PLACID: Identity-Preserving Multi-Object Compositing via Video Diffusion with Synthetic Trajectories
Gemma Canet Tarrés, Manel Baradad, Francesc Moreno-Noguer +1
Recent advances in generative AI have dramatically improved photorealistic image synthesis, yet they fall short for studio-level multi-object compositing. This task demands simulta…
Cost Savings from Automatic Quality Assessment of Generated Images
Xavier Giro-i-Nieto, Nefeli Andreou, Anqi Liang +3
Deep generative models have shown impressive progress in recent years, making it possible to produce high quality images with a simple text prompt or a reference image. However, st…
Separating Knowledge and Perception with Procedural Data
Adrián Rodríguez-Muñoz, Manel Baradad, Phillip Isola +1
We train representation models with procedural data only, and apply them on visual similarity, classification, and semantic segmentation tasks without further training by using vis…
A Vision Check-up for Language Models
Pratyusha Sharma, Tamar Rott Shaham, Manel Baradad +5
What does learning to model relationships between strings teach large language models (LLMs) about the visual world? We systematically evaluate LLMs' abilities to generate and reco…
Background Prompting for Improved Object Depth
Manel Baradad, Yuanzhen Li, Forrester Cole +4
Estimating the depth of objects from a single image is a valuable task for many vision, robotics, and graphics applications. However, current methods often fail to produce accurate…
Deep Augmentation: Dropout as Augmentation for Self-Supervised Learning
Rickard Brüel-Gabrielsson, Tongzhou Wang, Manel Baradad +1
Despite dropout's ubiquity in machine learning, its effectiveness as a form of data augmentation remains under-explored. We address two key questions: (i) When is dropout effective…
Procedural Image Programs for Representation Learning
Manel Baradad, Chun-Fu Chen, Jonas Wulff +4
Learning image representations using synthetic data allows training neural networks without some of the concerns associated with real images, such as privacy and bias. Existing wor…
Learning to See by Looking at Noise
Manel Baradad, Jonas Wulff, Tongzhou Wang +2
Current vision systems are trained on huge datasets, and these datasets come with costs: curation is expensive, they inherit human biases, and there are concerns over privacy and u…
Recurrent Neural Networks for Semantic Instance Segmentation
Amaia Salvador, Miriam Bellver, Victor Campos +4
We present a recurrent model for semantic instance segmentation that sequentially generates binary masks and their associated class probabilities for every object in an image. Our…