VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control

A gallery catalogue: light sans captions, Palatino essays and a thin rule between every room.

snap-research.github.io/vd3dVD3D: Taming Large Video Diffusion Transformers for 3D Camera Control
Screenshot of https://snap-research.github.io/vd3d/, © the authors. Shown for reference. Request removal
Venue
ICLR 2025
Year
2025
First author
Sherwin Bahmani
Field
video generation
Template
DreamFusion / MVDream template (Bootstrap 4.5)
Theme
Light

Checked against the rendered page. Headings are a light sans (Source Sans Pro 300) and only body paragraphs are Palatino serif; BibTeX is bare inline pink code.

Why it is here

Quiet, refined minimalism with light serif type, hairline section dividers, a framed teaser mosaic and a well-aligned three-column results layout. Shows what a narrow measure and generous whitespace achieve.

Overview

Color Palette

Typography

Source Sans Pro
VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control Abstract · Method · Results · BibTeX — 0123456789
Palatino Linotype
VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control Abstract · Method · Results · BibTeX — 0123456789

Spacing & Shape

Page Modules

Guidelines

More like this