diff --git a/assets/OmniLottie-data-pipeline.png b/assets/OmniLottie-data-pipeline.png new file mode 100644 index 0000000..892beff Binary files /dev/null and b/assets/OmniLottie-data-pipeline.png differ diff --git a/assets/OmniLottie-demo-1.mp4 b/assets/OmniLottie-demo-1.mp4 new file mode 100644 index 0000000..28068bc Binary files /dev/null and b/assets/OmniLottie-demo-1.mp4 differ diff --git a/assets/OmniLottie-demo-2.mp4 b/assets/OmniLottie-demo-2.mp4 new file mode 100644 index 0000000..b276f66 Binary files /dev/null and b/assets/OmniLottie-demo-2.mp4 differ diff --git a/assets/OmniLottie-demo-3.mp4 b/assets/OmniLottie-demo-3.mp4 new file mode 100644 index 0000000..5e42a46 Binary files /dev/null and b/assets/OmniLottie-demo-3.mp4 differ diff --git a/assets/OmniLottie-main-demo.mp4 b/assets/OmniLottie-main-demo.mp4 new file mode 100644 index 0000000..e3d0407 Binary files /dev/null and b/assets/OmniLottie-main-demo.mp4 differ diff --git a/assets/OmniLottie-pipeline.png b/assets/OmniLottie-pipeline.png new file mode 100644 index 0000000..633e045 Binary files /dev/null and b/assets/OmniLottie-pipeline.png differ diff --git a/assets/OmniLottie-t2lottie.png b/assets/OmniLottie-t2lottie.png new file mode 100644 index 0000000..5adca66 Binary files /dev/null and b/assets/OmniLottie-t2lottie.png differ diff --git a/assets/OmniLottie-teaser.png b/assets/OmniLottie-teaser.png new file mode 100644 index 0000000..fbc47b7 Binary files /dev/null and b/assets/OmniLottie-teaser.png differ diff --git a/assets/OmniLottie-ti2lottie.png b/assets/OmniLottie-ti2lottie.png new file mode 100644 index 0000000..1c71249 Binary files /dev/null and b/assets/OmniLottie-ti2lottie.png differ diff --git a/assets/OmniLottie-v2lottie.png b/assets/OmniLottie-v2lottie.png new file mode 100644 index 0000000..cea0f63 Binary files /dev/null and b/assets/OmniLottie-v2lottie.png differ diff --git a/index.html b/index.html index 9641391..68b947a 100644 --- a/index.html +++ b/index.html @@ -212,6 +212,11 @@ *Project Lead, Correspondence Authors + +
+ Accepted by CVPR 2026 +
+
1Fudan University, 2StepFun, @@ -317,7 +322,7 @@

Highlights

- OmniLottie Teaser + OmniLottie Teaser
@@ -334,7 +339,7 @@

Method Overview

- OmniLottie Pipeline + OmniLottie Pipeline
@@ -344,23 +349,6 @@
- - -
-
-

Lottie Tokenizer

-
-
- Lottie Tokenizer -
-
-
- Our Lottie tokenizer converts raw JSON files into compact sequences of shape, effect, and animation commands with their associated parameters. - This representation substantially shortens sequence lengths (81% reduction compared to raw JSON), improves token efficiency, and enables OmniLottie to learn a clean distribution over animation primitives. - The tokenizer supports five fundamental layer types: Precomposition, Solid, Null, Shape, and Text layers, each identified by a unique type parameter. -
-
-
@@ -374,19 +362,19 @@

Text-to-Lottie Generation

Text-Image-to-Lottie Generation

Video-to-Lottie Generation

@@ -415,7 +403,7 @@
- Text-to-Lottie Comparison + Text-to-Lottie Comparison
@@ -425,7 +413,7 @@
- Text-Image-to-Lottie Comparison + Text-Image-to-Lottie Comparison
@@ -435,7 +423,7 @@
- Video-to-Lottie Comparison + Video-to-Lottie Comparison
@@ -445,11 +433,6 @@

Quantitative Results

-
-
- Quantitative Results -
-
OmniLottie achieves 97.3%, 92.0%, and 90.7% success rates for Text-to-Lottie, Text-Image-to-Lottie, and Video-to-Lottie tasks respectively, substantially outperforming all baselines while producing richer token sequences that enable more expressive and detailed vector animation generation. @@ -461,11 +444,6 @@

MMLottie-2M Dataset

-
-
- MMLottie-2M Dataset -
-
The MMLottie-2M dataset consists of two million Lottie animations paired with multi-modal annotations including text descriptions, keyframe images, and rendered videos. Data is collected from major online platforms (LottieFiles, IconScout, Flaticon, Iconfont, Icons8) and augmented with SVG-derived animations. @@ -480,7 +458,7 @@

Data Processing Pipeline

- Data Processing Pipeline + Data Processing Pipeline