AI Paper Atlas

Research map / Video / Video generation

TIP-I2V: A Million-Scale Real Text and Image Prompt Dataset for Image-to-Video Generation

Wenhao Wang, Yi Yang

ICCV 2025 · Video › Video generation · approach: Compositional text-to-video diffusion · 9 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
train a VideoMAE video classifier instead of using frame-level fake image detectorsreal-vs-generated video detection · TIP-I2V detectionaccuracy: 53.1 -> 96.3Improved
evaluate detector cross-domain instead of same-domainVideoMAE classifier · TIP-I2V three-way classificationaccuracy: 96.3 -> 87.2Hurt
evaluate detector cross-domain instead of same-domainVideoMAE classifier · TIP-I2V three-way classificationmAP: 99.4 -> 94.8Hurt
train a CosFace metric-learning model for source tracing instead of pre-trained image copy detection modelsSSCD · TIP-TracemuAP: 77.5 -> 84.7Improved
test on I2VGen-XL cross-domain instead of same-domainCosFace baseline · TIP-TracemuAP: -17.0% muAPHurt
use real-user prompts (TIP-Eval) instead of expert-designed promptsPika vs CogVideoX-5B · TIP-Eval10 VBench dimensions: 8 out of 10Mixed
use CogVideoX-5B on calligraphy subject instead of cottageCogVideoX-5B · TIP-Evalaesthetic quality: 0.74 -> 0.40Hurt
use middle frame with fake image detectors on image-generated videosUnivFD and other detectors · TIP-I2V detectionaccuracy: 53.1No clear effect
compare existing image copy detection models on tracingSSCD · TIP-TracemuAP: 77.5No clear effect
compare video-text alignment across I2V modelsall five I2V models · TIP-Evalvideo-text alignment: 0.26No clear effect

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