.gradient-text { background: linear-gradient(to right, #0d47a1, #1976d2, #42a5f5, #90caf9); -webkit-background-clip: text; /* For WebKit browsers */ background-clip: text; /* Standard property */ color: transparent; /* Make the text color transparent to show gradient */ font-size: 100px; animation: gradient-wave 20s linear infinite; /* Smooth infinite animation */ background-size: 400% 100%; /* Larger gradient to ensure smooth looping */ }

• Objaverse-XL Release.

• GSO2, ABO, Toys4k, Pix3D, OmniObject3D Release.

• Shapenet Release.

• Objaverse 1.0 Release.

Annotation (Objaverse 1.0)

• id: Model identifier.
• filtered_name: Filtered Name (Using Mistral-Nemo).
• filtered_tags: Filtered Tags (Using Mistral-Nemo).
• filtered_description: Filtered Description (Using Mistral-Nemo).
• marvel_dense_description: Dense Multi-View Description (Generated by InternVL2-40M+).
• marvel_level_1: Comprehensive Annotation.
• marvel_level_2: Moderately Descriptive.
• marvel_level_3: Functional Semantic.
• marvel_level_4: Summary.
• marvel_level_5: Concise Tags.

Annotation (ShapeNet)

• id: Model identifier.
• synsetId: ShapeNet Synset Identifier.
• filtered_name: Filtered Name (Using Mistral-Nemo).
• filtered_tags: Filtered Tags (Using Mistral-Nemo).
• filtered_description: Filtered Description (Using Mistral-Nemo).
• marvel_dense_description: Dense Multi-View Description (Generated by InternVL2-40M+).
• marvel_level_1: Comprehensive Annotation.
• marvel_level_2: Moderately Descriptive.
• marvel_level_3: Functional Semantic.
• marvel_level_4: Summary.
• marvel_level_5: Concise Tags.

Annotation (GSO2, ABO, Toys4k, Pix3D, OmniObject3D)

• id: Model identifier.
• marvel_dense_description: Dense Multi-View Description (Generated by InternVL2-40M+).
• marvel_level_1: Comprehensive Annotation.
• marvel_level_2: Moderately Descriptive.
• marvel_level_3: Functional Semantic.
• marvel_level_4: Summary.
• marvel_level_5: Concise Tags.

If you use this dataset in your work, please cite the following publication.

@inproceedings{sinha2025marvel,
  title={MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation},
  author={Sinha, Sankalp and Khan, Mohammad Sadil and Usama, Muhammad and Sam, Shino and Stricker, Didier and Ali, Sk Aziz and Afzal, Muhammad Zeshan},
  booktitle={Proceedings of the Computer Vision and Pattern Recognition Conference},
  pages={8105--8116},
  year={2025}
}