New scene generation using advanced 3D Gaussian splatting
This study investigates a limitation of 3D Gaussian Splatting in real-time novel view synthesis and proposes a solution to address it. Despite its state-of-the-art performance, 3D Gaussian Splatting often results in unbalanced degrees of reconstruction, leading to blurred areas in rendered images (a...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175315 |
| _version_ | 1811678056166719488 |
|---|---|
| author | Wan, Yaxin |
| author2 | Lu Shijian |
| author_facet | Lu Shijian Wan, Yaxin |
| author_sort | Wan, Yaxin |
| collection | NTU |
| description | This study investigates a limitation of 3D Gaussian Splatting in real-time novel view synthesis and proposes a solution to address it. Despite its state-of-the-art performance, 3D Gaussian Splatting often results in unbalanced degrees of reconstruction, leading to blurred areas in rendered images (as depicted in Fig. 1). To address this issue, a progressive frequency control technique is implemented within the training process as inspired by FreeNeRF and FreGS. Specifically, Fourier transformation is applied
to both the resulting images and ground truth to map them into frequency spectrums, facilitating the learning of different frequency levels separately through the introduction of smooth filter masks. Experimental results on widely adopted benchmarks, including Mip-NeRF360, Tanks&Temples, and Deep Blending, demonstrate enhanced performance through the proposed frequency optimization technique. The study concludes that integrating frequency optimization into 3D Gaussian Splatting holds promise for
enhancing rendering quality. Moreover, recommendations for future research include exploring additional techniques such as compression and regularization to further enhance performance during the optimization process. |
| first_indexed | 2024-10-01T02:47:11Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/175315 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:47:11Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1753152024-04-26T15:44:44Z New scene generation using advanced 3D Gaussian splatting Wan, Yaxin Lu Shijian School of Computer Science and Engineering Shijian.Lu@ntu.edu.sg Computer and Information Science Engineering This study investigates a limitation of 3D Gaussian Splatting in real-time novel view synthesis and proposes a solution to address it. Despite its state-of-the-art performance, 3D Gaussian Splatting often results in unbalanced degrees of reconstruction, leading to blurred areas in rendered images (as depicted in Fig. 1). To address this issue, a progressive frequency control technique is implemented within the training process as inspired by FreeNeRF and FreGS. Specifically, Fourier transformation is applied to both the resulting images and ground truth to map them into frequency spectrums, facilitating the learning of different frequency levels separately through the introduction of smooth filter masks. Experimental results on widely adopted benchmarks, including Mip-NeRF360, Tanks&Temples, and Deep Blending, demonstrate enhanced performance through the proposed frequency optimization technique. The study concludes that integrating frequency optimization into 3D Gaussian Splatting holds promise for enhancing rendering quality. Moreover, recommendations for future research include exploring additional techniques such as compression and regularization to further enhance performance during the optimization process. Bachelor's degree 2024-04-23T06:07:05Z 2024-04-23T06:07:05Z 2024 Final Year Project (FYP) Wan, Y. (2024). New scene generation using advanced 3D Gaussian splatting. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175315 https://hdl.handle.net/10356/175315 en SCSE23-0097 application/pdf Nanyang Technological University |
| spellingShingle | Computer and Information Science Engineering Wan, Yaxin New scene generation using advanced 3D Gaussian splatting |
| title | New scene generation using advanced 3D Gaussian splatting |
| title_full | New scene generation using advanced 3D Gaussian splatting |
| title_fullStr | New scene generation using advanced 3D Gaussian splatting |
| title_full_unstemmed | New scene generation using advanced 3D Gaussian splatting |
| title_short | New scene generation using advanced 3D Gaussian splatting |
| title_sort | new scene generation using advanced 3d gaussian splatting |
| topic | Computer and Information Science Engineering |
| url | https://hdl.handle.net/10356/175315 |
| work_keys_str_mv | AT wanyaxin newscenegenerationusingadvanced3dgaussiansplatting |