AI Paper Atlas

Research map / Video / Video generation

GenDoP: Auto-regressive Camera Trajectory Generation as a Director of Photography

Mengchen Zhang, Tong Wu, Jing Tan, Ziwei Liu, Gordon Wetzstein, Dahua Lin

ICCV 2025 · Video › Video generation · approach: Trajectory-guided motion control · 19 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 canonical normalization to first-person tracking paths (ablation)GenDoP · DataDoP test (Motion)CLaTr-CLIP: 14.917 -> 36.179Improved
add canonical normalization (ablation)GenDoP · DataDoP test (Motion)CLaTr-FID: 68.590 -> 22.714Improved
add canonical normalization (ablation)GenDoP · DataDoP test (Motion)F1-score: 0.322 -> 0.400Improved
train the encoders instead of freezing them (ablation)GenDoP · DataDoP test (Motion)CLaTr-CLIP: 31.420 -> 36.179Improved
train the encoders instead of freezing them (ablation)GenDoP · DataDoP test (Motion)CLaTr-FID: 22.841 -> 22.714Improved
train Director3D on DataDoP instead of its pre-trained dataDirector3D · DataDoP test (Motion)CLaTr-CLIP: 0.000 -> 31.689Improved
use GenDoP auto-regressive framework instead of Director3D on the same DataDoP dataDirector3D (DataDoP) · DataDoP test (Motion)CLaTr-FID: 31.979 -> 22.714Improved
use GenDoP auto-regressive framework instead of Director3D on the same DataDoP dataDirector3D (DataDoP) · DataDoP test (Directorial)CLaTr-CLIP: 23.505 -> 32.408Improved
use GenDoP auto-regressive framework instead of Director3D on the same DataDoP dataDirector3D (DataDoP) · DataDoP test (Directorial)CLaTr-FID: 35.538 -> 34.275Improved
add first-frame RGBD input to textGenDoP · DataDoP testCLaTr-CLIP: 36.179 -> 30.231Hurt

Later papers that compared with this method

1 later accepted paper reports GenDoP as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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