AI Paper Atlas

Research map / Video / Video generation

DreamShot: Personalized Storyboard Synthesis with Video Diffusion Prior

Junjia Huang, Binbin Yang, Pengxiang Yan, Jiyang Liu, Bin Xia, Zhao Wang, Yitong Wang, Liang Lin, Guanbin Li

CVPR 2026 · Video › Video generation · approach: Trajectory-guided motion control · 0 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 role-attention consistency loss L_RAC (ablation)DreamShot · DreamShot Test Set Reference-to-ShotCIDS-Cross: 48.3 -> 51.6Improved
add L_RAC (ablation)DreamShot · DreamShot Test Set Reference-to-ShotAlignment score: 3.26 -> 3.39Improved
apply a phase offset to reference RoPE (ablation)DreamShot · Reference-to-ShotCIDS-Cross: 51.6 -> 20.0Hurt
apply a negative positional offset to reference (ablation)DreamShot · Reference-to-ShotCIDS-Cross: 51.6 -> 40.9Hurt
place reference image at sequence end instead of start (ablation)DreamShot · Reference-to-ShotCIDS-Cross: 51.6 -> 43.5Hurt
place reference image at sequence end instead of start (ablation)DreamShot · Reference-to-ShotAES: 5.50 -> 5.53Improved
use video-diffusion DreamShot instead of DreamODreamShot · Reference-to-Shot CIDS-CrossCIDS-Cross: 40.0 -> 51.6Improved
use DreamShot instead of second-best OmniGen2DreamShot · VistoryBenchCIDS-Cross: 54.8 -> 60.2Improved
train on real data only instead of real plus synthetic (ablation)DreamShot · Reference-to-ShotCIDS-Cross: 51.6 -> 52.2Mixed
train on synthetic data only instead of mixed (ablation)DreamShot · Reference-to-ShotAES: 5.50 -> 5.57Mixed
raise guidance omega1 to 8 (ablation)DreamShot · Reference-free CFGAES: omega1=1 -> omega1=8Hurt
raise lambda of L_RAC to 0.5 (ablation)DreamShot · Training objectiveAES: lambda 0.2 -> 0.5Hurt

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