Personalization has emerged as a prominent aspect within the field of
ge...
Recent advances in neural reconstruction enable high-quality 3D object
r...
Estimating the depth of objects from a single image is a valuable task f...
A critical obstacle preventing NeRF models from being deployed broadly i...
Image ad understanding is a crucial task with wide real-world applicatio...
Attaining a high degree of user controllability in visual generation oft...
Diffusion models, such as Stable Diffusion, have shown incredible perfor...
We present ShapeClipper, a novel method that reconstructs 3D object shap...
StyleGANs are at the forefront of controllable image generation as they
...
We present a method for joint alignment of sparse in-the-wild image
coll...
We present DreamBooth3D, an approach to personalize text-to-3D generativ...
Humans excel at acquiring knowledge through observation. For example, we...
Neural fields have emerged as a new paradigm for representing signals, t...
Machine learning models have been shown to inherit biases from their tra...
Creativity is an indispensable part of human cognition and also an inher...
Large-scale diffusion models have achieved state-of-the-art results on
t...
Universal Domain Adaptation (UniDA) deals with the problem of knowledge
...
Prompt tuning is a new few-shot transfer learning technique that only tu...
Large text-to-image models achieved a remarkable leap in the evolution o...
Deep long-tailed learning aims to train useful deep networks on practica...
Progress in GANs has enabled the generation of high-resolution photoreal...
The prime challenge in unsupervised domain adaptation (DA) is to mitigat...
Conventional domain adaptation (DA) techniques aim to improve domain
tra...
Inverse rendering of an object under entirely unknown capture conditions...
We present a novel 3D shape reconstruction method which learns to predic...
In this work, we introduce LEAD, an approach to discover landmarks from ...
Available 3D human pose estimation approaches leverage different forms o...
Articulation-centric 2D/3D pose supervision forms the core training obje...
The advances in monocular 3D human pose estimation are dominated by
supe...
Open compound domain adaptation (OCDA) has emerged as a practical adapta...
Decomposing a scene into its shape, reflectance and illumination is a
fu...
Single image 3D photography enables viewers to view a still image from n...
Unsupervised domain adaptation (DA) has gained substantial interest in
s...
Single image pose estimation is a fundamental problem in many vision and...
Remarkable progress has been made in 3D reconstruction of rigid structur...
Synthetic datasets play a critical role in pre-training CNN models for
o...
Prototype learning is extensively used for few-shot segmentation. Typica...
Representational learning hinges on the task of unraveling the set of
un...
We introduce the problem of perpetual view generation – long-range
gener...
Decomposing a scene into its shape, reflectance, and illumination is a
c...
End-to-end deep learning methods have advanced stereo vision in recent y...
Content creation, central to applications such as virtual reality, can b...
Point cloud registration is a fundamental problem in 3D computer vision,...
We present a self-supervised human mesh recovery framework to infer huma...
Neural rendering techniques promise efficient photo-realistic image synt...
Reconstructing 3D models from 2D images is one of the fundamental proble...
Camera captured human pose is an outcome of several sources of variation...
Training deep neural networks to estimate the viewpoint of objects requi...
Capturing the shape and spatially-varying appearance (SVBRDF) of an obje...
We learn a self-supervised, single-view 3D reconstruction model that pre...