Metraipolis › archive › dc_00bade89e60e

MSR: Multiple Subject Reference for Video Generation — arXiv:2609.18393

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18393
License (as archived)
CC0-1.0
Archived
2026-10-15T16:16:28.075Z
SHA-256 of the archived text
e61bb72ce17637641115f7dcdb647b69344013f367ca7002e39b0db2b9c463d8
Stable URL of this page
https://metraipolis.com/documento/dc_00bade89e60e

How to cite

MSR: Multiple Subject Reference for Video Generation — arXiv:2609.18393 — arXiv (en), https://arxiv.org/abs/2609.18393 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_00bade89e60e (Metraipolis archive dc_00bade89e60e, sha256 e61bb72ce17637641115f7dcdb647b69344013f367ca7002e39b0db2b9c463d8)
BibTeX
@misc{dc00bade89e60e,
  title = {MSR: Multiple Subject Reference for Video Generation — arXiv:2609.18393},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18393},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_00bade89e60e; sha256 e61bb72ce17637641115f7dcdb647b69344013f367ca7002e39b0db2b9c463d8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18393. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Conditioning a video generator on multiple images requires preserving appearance while associating each reference with its intended role. We present MSR (Multiple Subject Reference), a slot-aware conditioning scheme for LTX-based video generation. Each reference image is independently encoded as a static clip and represented by a separate latent-token group. A compact Fourier-feature multilayer perceptron adds a numeric slot embedding, while slot-dependent temporal offsets modify the group's rotary coordinates. The reference groups are prepended to noisy target tokens and serve as clean context during target-only flow-matching training. We implement this scheme through low-rank adaptation and release the resulting weights and inference workflows. Qualitative examples demonstrate compositions containing distinct characters and referenced environments in realistic and stylized scenes. Development observations suggest reduced reference confusion relative to an earlier continuous-reference baseline, while similar clothing, complex garments, and viewpoint changes remain challenging. We describe the conditioning mechanism, the retained training configuration, and the observed strengths and limitations of the released system. A supplementary audio-reference experiment adds voice conditioning while keeping the visual parameters frozen.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.18393 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T15:45:19.782Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).