Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
125 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

ECNR: Efficient Compressive Neural Representation of Time-Varying Volumetric Datasets (2311.12831v4)

Published 2 Oct 2023 in cs.CV, cs.GR, and cs.LG

Abstract: Due to its conceptual simplicity and generality, compressive neural representation has emerged as a promising alternative to traditional compression methods for managing massive volumetric datasets. The current practice of neural compression utilizes a single large multilayer perceptron (MLP) to encode the global volume, incurring slow training and inference. This paper presents an efficient compressive neural representation (ECNR) solution for time-varying data compression, utilizing the Laplacian pyramid for adaptive signal fitting. Following a multiscale structure, we leverage multiple small MLPs at each scale for fitting local content or residual blocks. By assigning similar blocks to the same MLP via size uniformization, we enable balanced parallelization among MLPs to significantly speed up training and inference. Working in concert with the multiscale structure, we tailor a deep compression strategy to compact the resulting model. We show the effectiveness of ECNR with multiple datasets and compare it with state-of-the-art compression methods (mainly SZ3, TTHRESH, and neurcomp). The results position ECNR as a promising solution for volumetric data compression.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. TTHRESH: Tensor compression for multidimensional visual data. IEEE TVCG, 26(9):2891–2903, 2020.
  2. Parametric correspondence and chamfer matching: Two new techniques for image matching. In Proc. IJCAI, pages 659–663, 1977.
  3. A generative model for volume rendering. IEEE TVCG, 25(4):1636–1650, 2019.
  4. Probabilistic data-driven sampling via multi-criteria importance analysis. IEEE TVCG, 27(12):4439–4454, 2021.
  5. Interactive spatio-temporal exploration of massive time-varying rectilinear scalar volumes based on a variable bit-rate sparse representation over learned dictionaries. C&G, 88:45–56, 2020.
  6. Homogeneity guided probabilistic data summaries for analysis and visualization of large-scale data sets. In Proc. PacificVis, pages 111–120, 2017.
  7. NeRVI: Compressive neural representation of visualization images for communicating volume visualization results. C&G, 116:216–227, 2023.
  8. J. Han and C. Wang. TSR-TVD: Temporal super-resolution for time-varying data analysis and visualization. IEEE TVCG, 26(1):205–215, 2020.
  9. J. Han and C. Wang. SSR-TVD: Spatial super-resolution for time-varying data analysis and visualization. IEEE TVCG, 28(6):2445–2456, 2022.
  10. J. Han and C. Wang. CoordNet: Data generation and visualization generation for time-varying volumes via a coordinate-based neural network. IEEE TVCG, 29(12):4951–4963, 2023.
  11. KD-INR: Time-varying volumetric data compression via knowledge distillation-based implicit neural representation. IEEE TVCG, 2023. Accepted.
  12. STNet: An end-to-end generative framework for synthesizing spatiotemporal super-resolution volumes. IEEE TVCG, 28(1):270–280, 2022.
  13. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. arXiv preprint arXiv:1510.00149, 2015.
  14. Learning both weights and connections for efficient neural network. In Proc. NeurIPS, pages 1135–1143, 2015.
  15. InSituNet: Deep image synthesis for parameter space exploration of ensemble simulations. IEEE TVCG, 26(1):23–33, 2020.
  16. D. A. Huffman. A method for the construction of minimum-redundancy codes. Proc. IRE, 40(9):1098–1101, 1952.
  17. Fast and effective lossy compression algorithms for scientific datasets. In Proc. ECPP, pages 843–856, 2012.
  18. Compressing the incompressible with ISABELA: In-situ reduction of spatio-temporal data. In Proc. ECPP, pages 366–379, 2011.
  19. Vox-Surf: Voxel-based implicit surface representation. IEEE TVCG, 2022. Accepted.
  20. Data reduction techniques for simulation, visualization and data analysis. CGF, 37(6):422–447, 2018.
  21. Error-controlled lossy compression optimized for high compression ratios of scientific datasets. In Proc. BigData, pages 438–447, 2018.
  22. SZ3: A modular framework for composing prediction-based error-bounded lossy compressors. IEEE TBD, 9(2):485–498, 2022.
  23. Dynamic model pruning with feedback. In Proc. ICLR, 2020.
  24. Compressive neural representations of volumetric scalar fields. CGF, 40(3):135–146, 2021.
  25. ACORN: Adaptive coordinate networks for neural scene representation. ACM ToG, 40(4):58:1–58:13, 2021.
  26. Importance estimation for neural network pruning. In Proc. CVPR, pages 11264–11272, 2019.
  27. Instant neural graphics primitives with a multiresolution hash encoding. ACM ToG, 41(4):102:1–102:15, 2022.
  28. S. Muraki. Approximation and rendering of volume data using wavelet transforms. In Proc. Vis, pages 21–22, 1992.
  29. Void-and-cluster sampling of large scattered data and trajectories. IEEE TVCG, 26(1):780–789, 2020.
  30. KiloNeRF: Speeding up neural radiance fields with thousands of tiny MLPs. In Proc. ICCV, pages 14315–14325, 2021.
  31. MINER: Multiscale implicit neural representation. In Proc. ECCV, pages 318–333, 2022.
  32. Implicit neural representations with periodic activation functions. In Proc. NeurIPS, pages 7462–7473, 2020.
  33. Topologically controlled lossy compression. In Proc. PacificVis, pages 46–55, 2018.
  34. TAMRESH–tensor approximation multiresolution hierarchy for interactive volume visualization. CGF, 32(3):151–160, 2013.
  35. Neural geometric level of detail: Real-time rendering with implicit 3D shapes. In Proc. CVPR, pages 11358–11367, 2021.
  36. K. Tang and C. Wang. STSR-INR: Spatiotemporal super-resolution for time-varying multivariate volumetric data via implicit neural representation. C&G, 2024. Accepted.
  37. C. Wang and J. Han. DL4SciVis: A state-of-the-art survey on deep learning for scientific visualization. IEEE TVCG, 29(8):3714–3733, 2023.
  38. C. Wang and K.-L. Ma. A statistical approach to volume data quality assessment. IEEE TVCG, 14(3):590–602, 2008.
  39. Statistical visualization and analysis of large data using a value-based spatial distribution. In Proc. PacificVis, pages 161–170, 2017.
  40. Fast neural representations for direct volume rendering. CGF, 41(6):196–211, 2022.
  41. Learning structured sparsity in deep neural networks. In Proc. NeurIPS, pages 2074–2082, 2016.
  42. HyperINR: A fast and predictive hypernetwork for implicit neural representations via knowledge distillation. arXiv preprint arXiv:2304.04188, 2023.
  43. Interactive volume visualization via multi-resolution hash encoding based neural representation. IEEE TVCG, 2023. Accepted.
  44. Deep hierarchical super resolution for scientific data. IEEE TVCG, 29(12):5483–5495, 2023.
  45. Adaptively placed multi-grid scene representation networks for large-scale data visualization. IEEE TVCG, 30(1):965–974, 2024.
  46. Neural fields in visual computing and beyond. CGF, 41(2):641–676, 2022.
  47. A unified framework of DNN weight pruning and weight clustering/quantization using ADMM. arXiv preprint arXiv:1811.01907, 2018.
  48. B.-L. Yeo and B. Liu. Volume rendering of DCT-based compressed 3D scalar data. IEEE TVCG, 1(1):29–43, 1995.
  49. The unreasonable effectiveness of deep features as a perceptual metric. In Proc. CVPR, pages 586–595, 2018.
Citations (4)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets