3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
We introduce an approach that creates animatable hu-man avatars from monocular videos using 3D Gaussian Splatting (3DGS). Existing methods based on neural radi-ance fields (NeRFs) achieve high-quality novel-viewlnovel-pose image synthesis but often require days of training, and are extremely slow at inference time. Recently, the com-munity has explored fast grid structures for efficient training of clothed avatars. Albeit being extremely fast at training, these methods can barely achieve an interactive ren-de ring frame rate with around 15 FPS. In this paper, we use 3D Gaussian Splatting and learn a non-rigid deformation network to reconstruct animatable clothed human avatars that can be trained within 30 minutes and rendered at real-time frame rates (50+ FPS). Given the explicit nature of our representation, we further introduce as-isometric-as-possible regularizations on both the Gaussian mean vectors and the covariance matrices, enhancing the generalization of our model on highly articulated unseen poses. Experi-mental results show that our method achieves comparable and even better performance compared to state-of-the-art approaches on animatable avatar creation from a monoc-ular input, while being 400x and 250x faster in training and inference, respectively. Please see our project page at https://neuralbodies.github.ioI3DGS-Avatar.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
3D Gaussian Splatting for Real-Time Radiance Field Rendering
Kerbl, B., Kopanas, G., Leimkuehler, T., & Drettakis, G. (2023) · ACM Transactions on Graphics · doi:10.1145/3592433 · Cited by 6,025
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
Mildenhall, B., Srinivasan, P. P., Tancik, M., Barron, J. T., Ramamoorthi, R., & Ng, R. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58452-8_24 · Cited by 5,351
Loper, M., Mahmood, N., Romero, J., Pons-Moll, G., & J, B. M. (2015) · ACM Transactions on Graphics · doi:10.1145/2816795.2818013 · Cited by 3,734 · cited 4×
“SMPL [26] parameter fittings from images can be inaccurate.”
BackgroundMethodInfluentialSCAPE: shape completion and animation of people
Anguelov, D., Srinivasan, P., Koller, D., Thrun, S., Rodgers, J., & Davis, J. (2005) · ACM Transactions on Graphics · doi:10.1145/1073204.1073207 · Cited by 1,489
AMASS: Archive of Motion Capture As Surface Shapes
Mahmood, N., Ghorbani, N., Troje, N. F., Pons-Moll, G., & Black, M. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00554 · Cited by 1,323
AI Choreographer: Music Conditioned 3D Dance Generation with AIST++
Li, R., Yang, S., Ross, D. A., & Kanazawa, A. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01315 · Cited by 553 · cited 1×
“We observe that our method preserves sharper details compared to ARAH and does not produce fluctuating artifacts as in HumanNeRF caused by noisy de-Table Additionally, we animate our learned avatars with pose sequences from AMASS [27] and AIST++ [20], shown in the rightmost column of Fig.”
MethodSTAR: Sparse Trained Articulated Human Body Regressor
Osman, A. A. A., Bolkart, T., & Black, M. J. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58539-6_36 · Cited by 252 · cited 1×
“To model human articulations, a widely adopted paradigm is to represent geometry and appearance in a shared canonical space [8, 12, 13, 22, 38, 41, 61, 62] Skinning (LBS) [2, 9, 26, 36, 37, 64] to deform the parametric human body under arbitrary poses.”
MethodGeometric modeling in shape space
Kilian, M., Mitra, N. J., & Pottmann, H. (2007) · ACM SIGGRAPH 2007 papers · doi:10.1145/1275808.1276457 · Cited by 224 · cited 2×
“Inspired by [44], we leverage the as-isometric-as-possible constraint [15] to restrict neighboring 3D Gaussian centers to preserve a similar distance after deformation.”
MethodNeural Articulated Radiance Field
Noguchi, A., Sun, X., Lin, S., & Harada, T. (2021) · arXiv · doi:10.48550/arxiv.2104.03110 · Cited by 177
Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering
Sitzmann, V., Rezchikov, S., Freeman, W. T., Tenenbaum, J. B., & Durand, F. (2021) · arXiv · doi:10.48550/arxiv.2106.02634 · Cited by 117 · cited 1×
“Recent advances in implicit neural fields [29, 33, 35, 39, 51, 54, 55, 57, 59, 69, 70] have enabled high-quality reconstruction of geometry [8, 41, 61, 65] and appearance [13, 22, 24, 34, 38, 40, 45, 62, 74] of clothed human bodies from sparse multi-view or monocular videos.”
Background
Metadata
- Authors
- Qian, Z., Wang, S., Mihajlovic, M., Geiger, A., & Tang, S.
- Venue
- 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2024-06-16
- Pages
- 5020-5030
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4402727627
- Cited by
- 175
- References
- 10
- FWCI
- 50.26
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.