Explanipedia

DuetSVG: Unified Multimodal SVG Generation with Internal Visual Guidance Open

Peiying Zhang, Nanxuan Zhao, Matthew Fisher, XU Yi-ran, Jing Liao , et al. · 2025

Recent vision-language model (VLM)-based approaches have achieved impressive results on SVG generation. However, because they generate only text and lack visual signals during decoding, they often struggle with complex semantics and fail t…

Rethinking Layered Graphic Design Generation with a Top-Down Approach Open

J.C. Chen, Zhaowen Wang, Ning Zhao, Li Zhang, Difan Liu , et al. · 2025

Graphic design is crucial for conveying ideas and messages. Designers usually organize their work into objects, backgrounds, and vectorized text layers to simplify editing. However, this workflow demands considerable expertise. With the ri…

How to Train Your Dragon: Automatic Diffusion‐Based Rigging for Characters with Diverse Topologies Open

Zeqi Gu, Difan Liu, Timothy Langlois, Matthew Fisher, Abe Davis · 2025

Recent diffusion‐based methods have achieved impressive results on animating images of human subjects. However, most of that success has built on human‐specific body pose representations and extensive training with labeled real videos. In …

How to Train Your Dragon: Automatic Diffusion-Based Rigging for Characters with Diverse Topologies Open

Zeqi Gu, Difan Liu, Timothy Langlois, Matthew Fisher, Abe Davis · 2025

Recent diffusion-based methods have achieved impressive results on animating images of human subjects. However, most of that success has built on human-specific body pose representations and extensive training with labeled real videos. In …

Mean-Shift Distillation for Diffusion Mode Seeking Open

Vikas Thamizharasan, Nikitas Chatzis, Iliyan Georgiev, Matthew Fisher, Evangelos Kalogerakis , et al. · 2025

We present mean-shift distillation, a novel diffusion distillation technique that provides a provably good proxy for the gradient of the diffusion output distribution. This is derived directly from mean-shift mode seeking on the distributi…

Move-in-2D: 2D-Conditioned Human Motion Generation Open

Hsin-Ping Huang, Yang Zhou, Jiayi Wang, Difan Liu, Feng Liu , et al. · 2024

Generating realistic human videos remains a challenging task, with the most effective methods currently relying on a human motion sequence as a control signal. Existing approaches often use existing motion extracted from other videos, whic…

Progressive Autoregressive Video Diffusion Models Open

Desai Xie, Zhan Xu, Yihang Hong, Hao Tan, Difan Liu , et al. · 2024

Current frontier video diffusion models have demonstrated remarkable results at generating high-quality videos. However, they can only generate short video clips, normally around 10 seconds or 240 frames, due to computation limitations dur…

HARIVO: Harnessing Text-to-Image Models for Video Generation Open

Mingi Kwon, Seoung Wug Oh, Yang Zhou, Difan Liu, Joon‐Young Lee , et al. · 2024

We present a method to create diffusion-based video models from pretrained Text-to-Image (T2I) models. Recently, AnimateDiff proposed freezing the T2I model while only training temporal layers. We advance this method by proposing a unique …

SNED: Superposition Network Architecture Search for Efficient Video Diffusion Model Open

Zhengang Li, Kang Yan, Yuchen Liu, Difan Liu, Tobias Hinz , et al. · 2024

While AI-generated content has garnered significant attention, achieving photo-realistic video synthesis remains a formidable challenge. Despite the promising advances in diffusion models for video generation quality, the complex model arc…

NIVeL: Neural Implicit Vector Layers for Text-to-Vector Generation Open

Vikas Thamizharasan, Difan Liu, Matthew Fisher, Nanxuan Zhao, Evangelos Kalogerakis , et al. · 2024

The success of denoising diffusion models in representing rich data distributions over 2D raster images has prompted research on extending them to other data representations, such as vector graphics. Unfortunately due to their variable str…

Attention-Driven Training-Free Efficiency Enhancement of Diffusion Models Open

Hongjie Wang, Difan Liu, Yan Kang, Yijun Li, Zhe Lin , et al. · 2024

Diffusion Models (DMs) have exhibited superior performance in generating high-quality and diverse images. However, this exceptional performance comes at the cost of expensive architectural design, particularly due to the attention module h…

VideoGigaGAN: Towards Detail-rich Video Super-Resolution Open

Yiran Xu, Taesung Park, Richard Zhang, Yang Zhou, Eli Shechtman , et al. · 2024

Video super-resolution (VSR) approaches have shown impressive temporal consistency in upsampled videos. However, these approaches tend to generate blurrier results than their image counterparts as they are limited in their generative capab…

Customize-A-Video: One-Shot Motion Customization of Text-to-Video Diffusion Models Open

Yixuan Ren, Yang Zhou, Jimei Yang, Jing Shi, Difan Liu , et al. · 2024

Image customization has been extensively studied in text-to-image (T2I) diffusion models, leading to impressive outcomes and applications. With the emergence of text-to-video (T2V) diffusion models, its temporal counterpart, motion customi…

Analysis of Spatiotemporal Changes in the Gravitational Structure of Urban Agglomerations in Northern and Southern Xinjiang Based on a Gravitational Model Open

Difan Liu, Yuejian Wang, Lei Wang, Liping Xu, Huanhuan Chen , et al. · 2023

The urban agglomeration plays a significant role in enhancing integrated regional development. Nevertheless, the expansion of urban agglomerations has demonstrated a lackluster ability to attract cities. Presently, finding solutions to sta…

VecFusion: Vector Font Generation with Diffusion Open

Vikas Thamizharasan, Difan Liu, Shantanu Agarwal, Matthew Fisher, Michaël Gharbi , et al. · 2023

We present VecFusion, a new neural architecture that can generate vector fonts with varying topological structures and precise control point positions. Our approach is a cascaded diffusion model which consists of a raster diffusion model f…

LRM: Large Reconstruction Model for Single Image to 3D Open

Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou , et al. · 2023

We propose the first Large Reconstruction Model (LRM) that predicts the 3D model of an object from a single input image within just 5 seconds. In contrast to many previous methods that are trained on small-scale datasets such as ShapeNet i…

Analysis of the Difference in Changes to Farmers’ Livelihood Capital under Different Land Transfer Modes—A Case Study of Manas County, Xinjiang, China Open

Difan Liu, Yuejian Wang, Yuejiao Chen, Guang Yang, Hailiang Xu , et al. · 2022

Farmers’ livelihoods alter as a direct result of land transfer. This study examined the impacts of land transfer on several indicators of farmers’ livelihood capital, as well as variations in the effects of different land transfer methods …

ASSET: Autoregressive Semantic Scene Editing with Transformers at High Resolutions Open

Difan Liu, Sandesh Shetty, Tobias Hinz, M. Fisher, Richard Zhang , et al. · 2022

We present ASSET, a neural architecture for automatically modifying an input high-resolution image according to a user's edits on its semantic segmentation map. Our architecture is based on a transformer with a novel attention mechanism. O…

Neural Strokes: Stylized Line Drawing of 3D Shapes Open

Difan Liu, Matthew Fisher, Aaron Hertzmann, Evangelos Kalogerakis · 2021

This paper introduces a model for producing stylized line drawings from 3D shapes. The model takes a 3D shape and a viewpoint as input, and outputs a drawing with textured strokes, with variations in stroke thickness, deformation, and colo…

Neural Shape Parsers for Constructive Solid Geometry Open

Gopal Sharma, Rishabh Goyal, Difan Liu, Evangelos Kalogerakis, Subhransu Maji · 2020

Constructive Solid Geometry (CSG) is a geometric modeling technique that defines complex shapes by recursively applying boolean operations on primitives such as spheres and cylinders. We present CSGNe, a deep network architecture that take…

Neural Contours: Learning to Draw Lines from 3D Shapes Open

Difan Liu, Mohamed Nabail, Aaron Hertzmann, Evangelos Kalogerakis · 2020

This paper introduces a method for learning to generate line drawings from 3D models. Our architecture incorporates a differentiable module operating on geometric features of the 3D model, and an image-based module operating on view-based …

ParSeNet: A Parametric Surface Fitting Network for 3D Point Clouds Open

Gopal Sharma, Difan Liu, Subhransu Maji, Evangelos Kalogerakis, Siddhartha Chaudhuri , et al. · 2020

CSGNet: Neural Shape Parser for Constructive Solid Geometry Open

Gopal Sharma, Rishabh Goyal, Difan Liu, Evangelos Kalogerakis, Subhransu Maji · 2018

We present a neural architecture that takes as input a 2D or 3D shape and outputs a program that generates the shape. The instructions in our program are based on constructive solid geometry principles, i.e., a set of boolean operations on…

Difan Liu YOU? Author Swipe