Learn More
We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convo-lutional networks and generative adversarial nets. The benefits of our model are threefold: first, the use of an adversarial(More)
In this paper, we investigate how, given an image, similar images sharing the same global description can help with unsupervised scene segmentation. In contrast to recent work in semantic alignment of scenes, we allow an input image to be explained by partial matches of similar scenes. This allows for a better explanation of the input scenes. We perform(More)
Humans demonstrate remarkable abilities to predict physical events in dynamic scenes, and to infer the physical properties of objects from static images. We propose a generative model for solving these problems of physical scene understanding from real-world videos and images. At the core of our generative model is a 3D physics engine, operating on an(More)
The Generic Viewpoint Assumption (GVA) states that the position of the viewer or the light in a scene is not special. Thus, any estimated parameters from an observation should be stable under small perturbations such as object, viewpoint or light positions. The GVA has been analyzed and quantified in previous works, but has not been put to practical use in(More)
The jump from problem sets to research can be hard. We sometimes see students who ace their classes struggle with their research. In little bites, here is what I think is important for succeeding in research as a graduate student. • The first advice can go on a bumper sticker: " Slow down to speed up ". In classes, the world is rigged. There's a simple(More)
Latent variables models have been widely applied in many problems in machine learning and related fields such as computer vision and information retrieval. However, the complexity of the latent space in such models is typically left as a free design choice. A larger latent space results in a more expressive model, but such models are prone to overfitting(More)
We present a nonparametric Bayesian method for texture learning and synthesis. A texture image is represented by a 2D Hidden Markov Model (2DHMM) where the hidden states correspond to the cluster labeling of textons and the transition matrix encodes their spatial layout (the compatibility between adjacent textons). The 2DHMM is coupled with the Hierarchical(More)