AI Paper Atlas

Research map / Video / Video generation

Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators

Levon Khachatryan, Andranik Movsisyan, Vahram Tadevosyan, Roberto Henschel, Zhangyang Wang, Shant Navasardyan, Humphrey Shi

ICCV 2023 · Video › Video generation · approach: Compositional text-to-video diffusion · 625 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add motion dynamics to initial latent codes (ablation)Text2Video-Zero · Qualitative temporal consistency (text-to-video, ControlNet edge)temporal consistency (qualitative)Improved
add cross-frame attention on first frame instead of self-attention (ablation)Text2Video-Zero · Qualitative identity preservationobject identity/appearance preservation (qualitative)Improved
combine motion latents with cross-frame attention (ablation)Text2Video-Zero · Qualitative temporal coherencetemporal coherence (qualitative)Improved
apply motion at T' < T (Delta t = 60) instead of Delta t = 0 (ablation)Text2Video-Zero · Qualitative object motionobject motion (qualitative)Improved
add background smoothing (alpha=0.6) (ablation)Text2Video-Zero · Qualitative background consistencybackground consistency (qualitative)Improved
replace per-frame Instruct-Pix2Pix with cross-frame attention versionVideo Instruct-Pix2Pix · Video editing consistencytemporal consistency (qualitative)Improved
apply method on top of ControlNetControlNet · Conditional video generationtemporal consistency (qualitative)Improved

Later papers that compared with this method

12 later accepted papers report Text2Video-Zero as a baseline in their results tables; 9 of them report a better number on most of the results they share with it.

4 more not listed.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy