ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, Byeongjun, Kim, Byung-Hoon, Chung, Hyungjin, Ye, Jong Chul
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914170068271104
author Park, Byeongjun
Kim, Byung-Hoon
Chung, Hyungjin
Ye, Jong Chul
author_facet Park, Byeongjun
Kim, Byung-Hoon
Chung, Hyungjin
Ye, Jong Chul
contents We present ReDirector, a novel camera-controlled video retake generation method for dynamically captured variable-length videos. In particular, we rectify a common misuse of RoPE in previous works by aligning the spatiotemporal positions of the input video and the target retake. Moreover, we introduce Rotary Camera Encoding (RoCE), a camera-conditioned RoPE phase shift that captures and integrates multi-view relationships within and across the input and target videos. By integrating camera conditions into RoPE, our method generalizes to out-of-distribution camera trajectories and video lengths, yielding improved dynamic object localization and static background preservation. Extensive experiments further demonstrate significant improvements in camera controllability, geometric consistency, and video quality across various trajectories and lengths.
format Preprint
id arxiv_https___arxiv_org_abs_2511_19827
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding
Park, Byeongjun
Kim, Byung-Hoon
Chung, Hyungjin
Ye, Jong Chul
Computer Vision and Pattern Recognition
We present ReDirector, a novel camera-controlled video retake generation method for dynamically captured variable-length videos. In particular, we rectify a common misuse of RoPE in previous works by aligning the spatiotemporal positions of the input video and the target retake. Moreover, we introduce Rotary Camera Encoding (RoCE), a camera-conditioned RoPE phase shift that captures and integrates multi-view relationships within and across the input and target videos. By integrating camera conditions into RoPE, our method generalizes to out-of-distribution camera trajectories and video lengths, yielding improved dynamic object localization and static background preservation. Extensive experiments further demonstrate significant improvements in camera controllability, geometric consistency, and video quality across various trajectories and lengths.
title ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.19827