An edit friendly ddpm noise space: Inversion and manipulations

I Huberman-Spiegelglas, V Kulikov… - Proceedings of the …, 2024 - openaccess.thecvf.com
Denoising diffusion probabilistic models (DDPMs) employ a sequence of white Gaussian
noise samples to generate an image. In analogy with GANs those noise maps could be …

Lepard: Learning explicit part discovery for 3d articulated shape reconstruction

D Liu, A Stathopoulos, Q Zhangli… - Advances in Neural …, 2024 - proceedings.neurips.cc
Reconstructing the 3D articulated shape of an animal from a single in-the-wild image is a
challenging task. We propose LEPARD, a learning-based framework that discovers …

Noiseclr: A contrastive learning approach for unsupervised discovery of interpretable directions in diffusion models

Y Dalva, P Yanardag - … of the IEEE/CVF Conference on …, 2024 - openaccess.thecvf.com
Generative models have been very popular in the recent years for their image generation
capabilities. GAN-based models are highly regarded for their disentangled latent space …

Deformer: Integrating transformers with deformable models for 3d shape abstraction from a single image

D Liu, X Yu, M Ye, Q Zhangli, Z Li… - Proceedings of the …, 2023 - openaccess.thecvf.com
Explicit 3D shape abstraction from a single 2D image is a long-standing problem in
computer vision and graphics. By leveraging a set of primitives to represent the target shape …

Posterior distillation sampling

J Koo, C Park, M Sung - … of the IEEE/CVF Conference on …, 2024 - openaccess.thecvf.com
Abstract We introduce Posterior Distillation Sampling (PDS) a novel optimization method for
parametric image editing based on diffusion models. Existing optimization-based methods …

Hidden Real Topology and Unusual Magnetoelectric Responses in Two‐Dimensional Antiferromagnets

J Gong, Y Wang, Y Han, Z Cheng, X Wang… - Advanced …, 2024 - Wiley Online Library
Recently, the real topology has been attracting widespread interest in two dimensions (2D).
Here, based on first‐principles calculations and theoretical analysis, we reveal the …

Lego: Learning egocentric action frame generation via visual instruction tuning

B Lai, X Dai, L Chen, G Pang, JM Rehg… - European Conference on …, 2025 - Springer
Generating instructional images of human daily actions from an egocentric viewpoint serves
as a key step towards efficient skill transfer. In this paper, we introduce a novel problem …

Stylegan-fusion: Diffusion guided domain adaptation of image generators

K Song, L Han, B Liu, D Metaxas… - Proceedings of the …, 2024 - openaccess.thecvf.com
Can a text-to-image diffusion model be used as a training objective for adapting a GAN
generator to another domain? In this paper, we show that the classifier-free guidance can be …

Source prompt disentangled inversion for boosting image editability with diffusion models

R Li, R Li, S Guo, L Zhang - European Conference on Computer Vision, 2025 - Springer
Text-driven diffusion models have significantly advanced the image editing performance by
using text prompts as inputs. One crucial step in text-driven image editing is to invert the …

Turboedit: Text-based image editing using few-step diffusion models

G Deutch, R Gal, D Garibi, O Patashnik… - SIGGRAPH Asia 2024 …, 2024 - dl.acm.org
Diffusion models have opened the path to a wide range of text-based image editing
frameworks. However, these typically build on the multi-step nature of the diffusion …