---
title: 社内勉強会_Gloria: Consistent Character Video Generation via Content Anchors
tags:  #画像生成ai #rope #gloria #社内勉強会 #輪読会 #cvpr2026 #video generation #3d rope #sca #superset content anchoring  
author: [NABLAS株式会社](https://www.docswell.com/user/nablas)
site: [Docswell](https://www.docswell.com/)
thumbnail: https://bcdn.docswell.com/page/4EM6KVD97W.jpg?width=480
description: 長尺動画生成においてキャラクターの一貫性を維持する手法「Gloria」の解説資料です。視点や表情のアンカーフレームと3D RoPEを活用し、長時間でも顔や姿の崩れない安定した動画生成を実現します。  This presentation introduces "Gloria," a method for maintaining character consistency in long-form video generation. By leveraging content anchors and 3D RoPE, it achieves stable character generation across long durations, multiple viewpoints, and expressions.
published: October 26, 26
canonical: https://www.docswell.com/s/nablas/K3JV6J-2026-10-26-161815
---
# Page. 1

![Page Image](https://bcdn.docswell.com/page/4EM6KVD97W.jpg)

Gloria: Consistent Character Video
Generation via Content Anchors
CVPR 2026


# Page. 2

![Page Image](https://bcdn.docswell.com/page/PERPX24979.jpg)

Background
◆ Digital characters
High-quality static character
images
Long-form production
Short clips generation
Image generation
→
Motion and temporal coherence
in short clips
→
Extended narratives and
reusable digital characters
Generative AI is moving from isolated visual content to scalable, character-driven storytelling.
➢ Commercial applications
- live streaming
- virtual influencer
- episodic series


# Page. 3

![Page Image](https://bcdn.docswell.com/page/P7X36D23EX.jpg)

Background
◆ Three key goals of character video generation
Long duration
Generation
Multi-view appearance
consistency
Expressive identity
consistency
Generate extended video
sequences while
maintaining character
identity over time
Preserve a character’s
appearance across different
viewpoints and camera
angles
Maintain the same character
identity across different facial
expressions and emotions
Across time
Across viewpoints
Across expressions


# Page. 4

![Page Image](https://bcdn.docswell.com/page/37KY1XMNED.jpg)

Motivation
◆ Existing problems
Insufficient Identity Context
Non-character-centric
Memory
Limited reference images cannot fully
capture a character’s appearance
across different viewpoints and facial
expressions.
Historical video frames contain
redundant motion and scene
information, rather than explicitly
preserving stable character identity.
Both limitations lead to identity drift and inconsistent character appearance.
How to provide stable and comprehensive character information throughout long-form
video generation?


# Page. 5

![Page Image](https://bcdn.docswell.com/page/LJ39Q5YZJ5.jpg)

Proposal
◆ Character-centric content anchors
- represent character visual attributes using a compact set of anchor frames
- provide persistent visual references to maintain consistency throughout long-form video generation
Global anchor
(Cg)
Viewpoint anchors
(Cv)
Expression
anchors (Ce)
Preserve global scene context
and subject–background
relationship
Capture character appearance
from different viewpoints
(front / back / left / right…)
Capture character identity
across different facial
expressions. (neutral / happy /
angry / sad...)
Long-term consistency
Multi-view consistency
Expressive identity consistency


# Page. 6

![Page Image](https://bcdn.docswell.com/page/8JDG9D54JG.jpg)

Proposal
◆ Challenges of reference-based video generation
1. Copy-pasting
2. Multi-reference Conflicts
The model may directly reproduce
reference frames instead of
learning transferable character
attributes.
Different anchors may interfere with
one another, making it difficult to
distinguish and use the right
reference information.
How to use multiple content anchors without copying or interference?


# Page. 7

![Page Image](https://bcdn.docswell.com/page/VEP3ZXGV78.jpg)

Method
◆ Mechanism 1: SCA
Challenge 1: Copy-pasting
Training clips
Superset content anchoring
(SCA)
Content anchors
Sample anchors from inside and
outside the target training clip to
discourage direct duplication.
Whole video
SCA: expand anchor sampling to learn
transferable character features


# Page. 8

![Page Image](https://bcdn.docswell.com/page/27V4WL6REQ.jpg)

Method
◆ Mechanism 2: RoPE
Challenge 2: Multi-reference conflicts
Rotary Position Embedding
(RoPE)
Assign distinct positional offsets to
anchors so the model can distinguish
their roles.
Video Tokens → normal positions
Global Anchor → offset 600
Viewpoint Anchors → offset 200
Expression Anchors → offset 400


# Page. 9

![Page Image](https://bcdn.docswell.com/page/5JG1DXN67L.jpg)

Method
◆ Overall pipeline
1.
2.
3.
4.
Anchor sampling & Encoding: sample character-centric anchors and encode them into latent space
Anchor differentiation: assign distinct RoPE positional offsets to distinguish anchor types and subcategories
Multimodal conditioning: integrate content anchors with text, image, and audio conditions through attention mechanisms
Decoding & Video generation: denoise video latents and decode them into video frames


# Page. 10

![Page Image](https://bcdn.docswell.com/page/47QD18Q2EP.jpg)

Experiments
◆ Setup
➢ Dataset
• Large-scale single-person video clips
• Training data covering diverse character appearances, motions, and expressions
➢ Evaluation metrics
• Long-term: subject consistency, background consistency, ArcFace
• Multi-view: DINO-I, CLIP-I
• Expressive: Exp.


# Page. 11

![Page Image](https://bcdn.docswell.com/page/KE4ZR2NPJ1.jpg)

Experiments
◆ Results: long-term consistency
Method
Subject ↑
Background ↑
ArcFace ↑
HunyuanAvatar
0.914
0.909
0.495
WanS2V
0.872
0.864
0.439
InfiniteTalk
0.952
0.942
0.695
Gloria
0.960
0.951
0.787
Gloria achieves the best long-term subject, background, and facial identity consistency.


# Page. 12

![Page Image](https://bcdn.docswell.com/page/L71RQM5X7G.jpg)

Experiments
◆ Visualization results


# Page. 13

![Page Image](https://bcdn.docswell.com/page/G7W1M35KJ2.jpg)

Experiments
◆ Results: multi-view and expressive ID
Method
DINO-I ↑
CLIP-I ↑
Exp. ↑
Phantom
0.597
0.785
0.645
Vidu
0.662
0.822
0.644
Kling
0.698
0.797
0.599
Humo
0.810
0.816
0.672
Gloria
0.821
0.858
0.717
Gloria outperforms competing methods in appearance similarity and expressive identity consistency.


# Page. 14

![Page Image](https://bcdn.docswell.com/page/4JZPQVNN73.jpg)

Experiments
◆ Ablation study
Long-term Consistency
Setting
Subject ↑
Background ↑
ArcFace ↑
Full Gloria
0.960
0.951
0.787
w/o Global Anchor
0.840
0.844
0.233
w/o SCA
0.922
0.899
0.489
Setting
DINO-I ↑
CLIP-I ↑
Exp. ↑
Full Gloria
0.821
0.858
0.717
w/o Viewpoint / Expression Anchors
0.801
0.848
0.668
w/o RoPE
0.817
0.855
0.671
Appearance & Expression Consistency


# Page. 15

![Page Image](https://bcdn.docswell.com/page/YE6M6589JV.jpg)

Summary
Limitations
Goal
Method
Long-form character
consistency
Character-centric
content anchors
Limited
Generalization
Maintain character identity
across long video
sequences, different
viewpoints, and facial
expressions.
Use global, viewpoint, and
expression anchors to
provide stable visual
references.
Primarily focuses on singleperson videos, with limited
support for multi-character
interactions.
SCA reduces copy-pasting,
while RoPE helps distinguish
different anchors.
Requires diverse reference
frames and substantial
training resources.


# Page. 16

![Page Image](https://bcdn.docswell.com/page/GE5Z6Y9D74.jpg)

Thank you!
Q&A