室外大场景神经辐射场综述

doi:10.11996/JG.j.2095-302X.2024040631

摘要/Abstract

摘要：

对室外大场景进行三维建模，不仅可以完成实时的城市建图和漫游，还能为自动驾驶等提供技术支持。近年来，神经隐式建模的发展十分迅速，神经辐射场(NeRF)的出现更是将神经隐式建模推上了新的高度。NeRF凭借高质量渲染和任意角度渲染的特点，已经在可控编辑、数字化人体、城市场景重建等领域得到了广泛的应用。NeRF通过深度学习的方法从二维图片及其位姿中学习隐式三维场景，生成新视角图像。然而原始NeRF只能在有界场景下得到逼真的结果，在对室外大场景进行建模时，往往会面临无界背景、模型容量、场景外观等问题。基于室外大场景中NeRF部署的多个角度改进和多样的NeRF变种，首先介绍NeRF的背景，然后从室外大场景的难点切入，对各个难点的解决方法进行分析和讨论，最后对室外大场景NeRF目前的进展进行总结并对未来进行展望。

关键词: 神经隐式表达, 神经辐射场, 新视角合成, 室外场景, 大场景

Abstract:

The 3D modeling of large outdoor scenes can not only complete real-time urban mapping and roaming, but also provide technical support for autonomous driving. In recent years, the advancement of neural implicit representation has been rapid, and the emergence of the neural radiance fields (NeRF) has propelled neural implicit representation to a new height. With its characteristics of high-quality rendering and arbitrary angle rendering, NeRF has been widely applied in controllable editing, digital human body, urban scene reconstruction, and other fields. The neural radiance field utilizes deep learning methods to learn implicit three-dimensional scenes from two-dimensional pictures and their poses, synthesizing novel view images.However, the original NeRF can only yield realistic results in bounded scenes, posing challenges in modeling large outdoor scenes due to problems such as unbounded backgrounds, model capacity constraints, and scene appearance. In order to deploy NeRF in large outdoor scenes, researchers have improved from multiple angles and proposed a variety of NeRF variants. Our review will begin by introducing the background of neural radiance fields, then delve into the challenges specific to the large outdoor scenes, analyzing and discussing the solutions to each, before concluding with a summary of current progress of NeRF for large outdoor scenes and prospects for the future.

Key words: neural implicit representation, neural radiance field, novel view synthesis, outdoor scenes, large scenes

中图分类号:

TP391

董相涛, 马鑫, 潘成伟, 鲁鹏. 室外大场景神经辐射场综述[J]. 图学学报, 2024, 45(4): 631-649.

DONG Xiangtao, MA Xin, PAN Chengwei, LU Peng. A review of neural radiance fields for outdoor large scenes[J]. Journal of Graphics, 2024, 45(4): 631-649.

图/表 24

图1 NeRF算法流程[5]

Fig. 1 The pipeline of NeRF[5]

图2 NeRF总体研究框架

Fig. 2 The NeRF overall research framework

表1 室外大场景NeRF研究工作总结

Table 1 Summary of NeRF research work in outdoor large scenes

难点	代表文献	方法分类
无界问题	文献[50,80]	引入反向球体参数化
	文献[81-82]	引入非线性场景参数化
	文献[83]	引入透视转换
大场景建模问题	文献[1,50,84]	引入分解思想
	文献[85]	应用网格化
	文献[2,86]	使用了深度监督思想
场景外观与动态物体问题	文献[87-88]	建模外观
场景外观与动态物体问题	文献[44-46]	对动态物体进行建模
模型的泛化性	文献[28-29,47]	引入卷积神经网络
模型的泛化性	文献[48-49]	引入极线信息

图3 用1/r表示(x', y', z')[80]

Fig. 3 Use 1/r to represent (x', y', z')[80]

图4 NeRF ++对比Mega-NeRF[50]

Fig. 4 Comparison of NeRF++ with Mega-NeRF[50] ((a) NeRF++; (b) Mega-NeRF)

图5 Mip-NeRF采样示意图[14]

Fig. 5 The diagram of Mip-NeRF sampling[14]

图6 Mip-NeRF 360算法流程[81]

Fig. 6 The pipeline of Mip-NeRF 360[81]

图7 Mip- NeRF 360和MERF对比[82]

Fig. 7 Comparison of Mip-NeRF 360 and MERF[82] ((a) Contract(?); (b) Contractπ(?))

图8 Instant-NGP算法流程[20]

Fig. 8 The pipeline of Instant-NGP[20]

图9 F2-NeRF算法流程[83] ((a)空间划分；(b)局部转换；(c)不同的哈希正数；(d)同一哈希正数索引)

Fig. 9 The pipeline of F2-NeRF[83] ((a) Space subdivision; (b) Local warping; (c) Different Hash function; (d) Indexed with same Hash table)

表2 室外数据集上4种解决无界问题方法的性能[82-83]

Table 2 Performance of three methods for solving unbounded problems on a outdoor dataset[82-83]

方法	PSNR↑	SSIM↑	LPIPS↓
NeRF++	23.47	0.603	0.499
Mip-NeRF360	27.01	0.766	0.295
Mip-NeRF360(short)	22.04	0.537	0.586
MERF	23.19	0.616	0.343
F^2-NeRF	26.32	0.779	0.276

图10 Block-NeRF模型结构[1]

Fig. 10 The model structure of Block-NeRF[1]

图11 Mega-NeRF，Block-NeRF和Switch-NeRF对比[87]

Fig. 11 Comparison of Mega-NeRF, Block-NeRF and Switch-NeRF[87] ((a) Mega-NeRF; (b) Block-NeRF; (c) Switch-NeRF)

图12 Switch-NeRF算法流程[87]

Fig. 12 The pipeline of Switch-NeRF[87]

表3 航拍建筑数据集上Mega-NeRF和Switch-NeRF的性能比较[84]

Table 3 Performance comparison of Mega-NeRF and Switch-NeRF on aerial building datasets[84]

方法	PSNR↑	SSIM↑	LPIPS↓
Mega-NeRF	20.93	0.547	0.504
Switch-NeRF	21.54	0.579	0.474

表4 航拍校园数据集上Mega-NeRF和Switch-NeRF的性能比较[84]

Table 4 Performance comparison of Mega-NeRF and Switch-NeRF on aerial campus datasets[84]

方法	PSNR↑	SSIM↑	LPIPS↓
Mega-NeRF	23.42	0.537	0.618
Switch-NeRF	23.62	0.541	0.609

图13 Grid-guided NeRF算法流程[88] ((a)两阶段训练；(b)两分支模型设计)

Fig. 13 The pipeline of Grid-guided NeRF[88] ((a) Two stage training scheme; (b) Two branch model design)

图14 DS-NeRF算法流程[32]

Fig. 14 The pipeline of DS-NeRF[32]

图15 Urban-NeRF算法结果[2]

Fig. 15 The result of Urban-NeRF[2]

表5 在Waymo数据集上S-NeRF和Urban-NeRF的性能比较[86]

Table 5 Comparison of the performance of S-NeRF and Urban-NeRF on the Waymo dataset[86]

方法	PSNR↑	SSIM↑	LPIPS↓
S-NeRF	23.60	0.743	0.422
Urban-NeRF	17.80	0.494	0.701

图16 SUDS算法流程[46] ((a)体素查找；(b)索引；(c) MLP预测；(d)输出融合)

Fig. 16 The pipeline of SUDS[46] ((a) Voxel lookup; (b) Indexing; (c) MLP evaluation; (d) Output blending)

图17 PixelNeRF算法流程[29]

Fig. 17 The pipeline of PixelNeRF[29]

图18 MVS-NeRF算法流程[28] ((a)代价体积；(b)神经编码体；(c)体渲染)

Fig. 18 The pipeline of MVS-NeRF[28] ((a) Cost volume; (b) Neural encoding volume; (c) Volume render)

表6 在DTU数据集上5种泛化性方法性能的比较[48-49]

Table 6 Comparison of performance of five generalization methods on the DTU dataset[48-49]

方法	PSNR↑	SSIM↑	LPIPS↓
PixelNeRF	19.31	0.789	0.671
IBRNet	26.04	0.917	0.190
MVSNeRF	26.63	0.931	0.168
EVE-NeRF	27.80	0.937	0.149
PBNR	28.50	0.932	0.167

参考文献 92

[1]	TANCIK M, CASSER V, YAN X C, et al. Block-NeRF: scalable large scene neural view synthesis[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 8238-8248.
[2]	REMATAS K, LIU A, SRINIVASAN P, et al. Urban radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12922-12932.
[3]	LI Z P, LI L, ZHU J K. READ: large-scale neural scene rendering for autonomous driving[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(2): 1522-1529.
[4]	ZHU Z H, PENG S Y, LARSSON V, et al. NICE-SLAM: neural implicit scalable encoding for SLAM[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12776-12786.
[5]	MILDENHALL B, SRINIVASAN P P, TANCIK M, et al. NeRF: representing scenes as neural radiance fields for view synthesis[M]//Computer Vision - ECCV 2020. Cham: Springer International Publishing, 2020: 405-421.
[6]	CHEN Y B, LIU S F, WANG X L. Learning continuous image representation with local implicit image function[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 8624-8634.
[7]	XU X Q, WANG Z Y, SHI H. UltraSR: spatial encoding is a missing key for implicit image function-based arbitrary-scale super-resolution[EB/OL]. [2023-09-10]. https://arxiv.org/abs/2103.12716.
[8]	SKOROKHODOV I, IGNATYEV S, ELHOSEINY M. Adversarial generation of continuous images[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 10748-10759.
[9]	SHAHAM T R, GHARBI M, ZHANG R, et al. Spatially-adaptive pixelwise networks for fast image translation[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 14877-14886.
[10]	CHEN Z Q, ZHANG H. Learning implicit fields for generative shape modeling[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2019: 5932-5941.
[11]	MESCHEDER L, OECHSLE M, NIEMEYER M, et al. Occupancy networks: learning 3D reconstruction in function space[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2019: 4455-4465.
[12]	PARK J J, FLORENCE P, STRAUB J, et al. DeepSDF: learning continuous signed distance functions for shape representation[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2019: 165-174.
[13]	SUBAKAN C, RAVANELLI M, CORNELL S, et al. Attention is all you need in speech separation[C]// ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing. New York: IEEE Press, 2021: 21-25.
[14]	BARRON J T, MILDENHALL B, TANCIK M, et al. Mip-NeRF: a multiscale representation for anti-aliasing neural radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5835-5844.
[15]	VERBIN D, HEDMAN P, MILDENHALL B, et al. Ref-NeRF: structured view-dependent appearance for neural radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 5481-5490.
[16]	PARK K, SINHA U, HEDMAN P, et al. HyperNeRF: a higher-dimensional representation for topologically varying neural radiance fields[J]. ACM Transactions on Graphics, 40(6): 238:1-238:12.
[17]	MILDENHALL B, HEDMAN P, MARTIN-BRUALLA R, et al. NeRF in the dark: high dynamic range view synthesis from noisy raw images[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 16169-16178.
[18]	LIU L J, GU J T, LIN K Z, et al. Neural sparse voxel fields[EB/OL]. [2023-09-10]. https://arxiv.org/abs/2007.11571.
[19]	LINDELL D B, MARTEL J N P, WETZSTEIN G. AutoInt: automatic integration for fast neural volume rendering[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 14551-14560.
[20]	MÜLLER T, EVANS A, SCHIED C, et al. Instant neural graphics primitives with a multiresolution hash encoding[J]. ACM Transactions on Graphics, 2022, 41(4): 1-15.
[21]	HEDMAN P, SRINIVASAN P P, MILDENHALL B, et al. Baking neural radiance fields for real-time view synthesis[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5855-5864.
[22]	YU A, LI R L, TANCIK M, et al. PlenOctrees for real-time rendering of neural radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5732-5741.
[23]	GARBIN S J, KOWALSKI M, JOHNSON M, et al. FastNeRF: high-fidelity neural rendering at 200FPS[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 14326-14335.
[24]	REISER C, PENG S Y, LIAO Y Y, et al. KiloNeRF: speeding up neural radiance fields with thousands of tiny MLPs[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 14315-14325.
[25]	FRIDOVICH-KEIL S, YU A, TANCIK M, et al. Plenoxels: radiance fields without neural networks[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 5491-5500.
[26]	SUN C, SUN M, CHEN H T. Direct voxel grid optimization: super-fast convergence for radiance fields reconstruction[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 5449-5459.
[27]	CHEN A P, XU Z X, GEIGER A, et al. TensoRF: tensorial radiance fields[C]// The 17th European Conference on Computer Vision. Cham: Springer2022: 333-350.
[28]	CHEN A P, XU Z X, ZHAO F Q, et al. MVSNeRF: fast generalizable radiance field reconstruction from multi-view stereo[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 14104-14113.
[29]	YU A, YE V, TANCIK M, et al. pixelNeRF: neural radiance fields from one or few images[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 4576-4585.
[30]	LIU Y, PENG S D, LIU L J, et al. Neural rays for occlusion-aware image-based rendering[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 7814-7823.
[31]	JAIN A, TANCIK M, ABBEEL P. Putting NeRF on a diet: semantically consistent few-shot view synthesis[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5865-5874.
[32]	DENG K L, LIU A, ZHU J Y, et al. Depth-supervised NeRF: fewer views and faster training for free[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12872-12881.
[33]	NIEMEYER M, GEIGER A. GIRAFFE: representing scenes as compositional generative neural feature fields[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 11448-11459.
[34]	SCHWARZ K, LIAO Y Y, NIEMEYER M, et al. GRAF: generative radiance fields for 3D-aware image synthesis[EB/OL]. [2023-09-10]. https://arxiv.org/abs/2007.02442.
[35]	MENG Q, CHEN A P, LUO H M, et al. GNeRF: GAN-based neural radiance field without posed camera[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 6331-6341.
[36]	KOSIOREK A R, STRATHMANN H, ZORAN D, et al. NeRF-VAE: a geometry aware 3D scene generative model[EB/OL]. [2023-09-10]. http://arxiv.org/abs/2104.00587.
[37]	LIU S, ZHANG X M, ZHANG Z T, et al. Editing conditional radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5753-5763.
[38]	WANG C, CHAI M L, HE M M, et al. CLIP-NeRF: text-and-image driven manipulation of neural radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 3825-3834.
[39]	WANG R B, ZHANG S, HUANG P, et al. Semantic is enough: only semantic information for NeRF reconstruction[C]// 2023 IEEE International Conference on Unmanned Systems. New York: IEEE Press, 2023: 906-912.
[40]	KUNDU A, GENOVA K, YIN X Q, et al. Panoptic neural fields: a semantic object-aware neural scene representation[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12861-12871.
[41]	SUCAR E, LIU S K, ORTIZ J, et al. iMAP: implicit mapping and positioning in real-time[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 6209-6218.
[42]	LIN C H, MA W C, TORRALBA A, et al. BARF: bundle-adjusting neural radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5721-5731.
[43]	JEONG Y, AHN S, CHOY C, et al. Self-calibrating neural radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5826-5834.
[44]	PUMAROLA A, CORONA E, PONS-MOLL G, et al. D-NeRF: neural radiance fields for dynamic scenes[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 10313-10322.
[45]	FANG J M, YI T R, WANG X G, et al. Fast dynamic radiance fields with time-aware neural voxels[EB/OL]. [2023-09-10]. https://arxiv.org/pdf/2205.15285.pdf..
[46]	TURKI H, ZHANG J Y, FERRONI F, et al. SUDS: scalable urban dynamic scenes[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2023: 12375-12385.
[47]	WANG Q Q, WANG Z C, GENOVA K, et al. IBRNet: learning multi-view image-based rendering[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 4688-4697.
[48]	SUHAIL M, ESTEVES C, SIGAL L, et al. Generalizable patch-based neural rendering[EB/OL]. [2023-09-10]. http://arxiv.org/abs/2207.10662.
[49]	MIN Z Y, LUO Y W, YANG W, et al. Entangled view-epipolar information aggregation for generalizable neural radiance fields[EB/OL]. [2023-09-10]. https://arxiv.org/abs/2311.11845.
[50]	TURKI H, RAMANAN D, SATYANARAYANAN M. Mega-NeRF: scalable construction of large-scale NeRFs for virtual fly- throughs[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12912-12921.
[51]	XIANGLI Y B, XU L N, PAN X G, et al. BungeeNeRF: progressive neural radiance field for extreme multi-scale scene rendering[C]// The 17th European Conference on Computer Vision. Cham: Springer, 2022: 106-122.
[52]	JANG W, AGAPITO L. CodeNeRF: disentangled neural radiance fields for object categories[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 12929-12938.
[53]	KANIA K, YI K M, KOWALSKI M, et al. CoNeRF: controllable neural radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 18602-18611.
[54]	XIE C, PARK K, MARTIN-BRUALLA R, et al. FiG-NeRF: figure-ground neural radiance fields for 3D object category modelling[C]// 2021 International Conference on 3D Vision (3DV). New York: IEEE Press, 2021: 962-971.
[55]	MA L, LI X Y, LIAO J, et al. Deblur-NeRF: neural radiance fields from blurry images[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 12851-12860.
[56]	HUANG X, ZHANG Q, FENG Y, et al. HDR-NeRF: high dynamic range neural radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 18377-18387.
[57]	LI Z S, MÜLLER T, EVANS A, et al. Neuralangelo: high-fidelity neural surface reconstruction[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2023: 8456-8465.
[58]	AZINOVIĆ D, MARTIN-BRUALLA R, GOLDMAN D B, et al. Neural RGB-D surface reconstruction[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 6280-6291.
[59]	OECHSLE M, PENG S Y, GEIGER A. UNISURF: unifying neural implicit surfaces and radiance fields for multi-view reconstruction[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5569-5579.
[60]	PARK K, SINHA U, BARRON J T, et al. Nerfies: deformable neural radiance fields[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5845-5854.
[61]	HONG Y, PENG B, XIAO H Y, et al. HeadNeRF: a realtime NeRF-based parametric head model[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 20342-20352.
[62]	PENG S D, ZHANG Y Q, XU Y H, et al. Neural body: implicit neural representations with structured latent codes for novel view synthesis of dynamic humans[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 9050-9059.
[63]	WENG C, CURLESS B, SRINIVASAN P P, et al. HumanNeRF: free-viewpoint rendering of moving people from monocular video[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 16189-16199.
[64]	SHAO R Z, ZHANG H W, ZHANG H, et al. DoubleField: bridging the neural surface and radiance fields for high-fidelity human reconstruction and rendering[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 15851-15861.
[65]	ZHI Y H, QIAN S H, YAN X H, et al. Dual-space NeRF: learning animatable avatars and scene lighting in separate spaces[C]// 2022 International Conference on 3D Vision. New York: IEEE Press, 2022: 1-10.
[66]	GAO K, GAO Y N, HE H J, et al. NeRF: neural radiance field in 3D vision, a comprehensive review[EB/OL]. [2023-09-10]. http://arxiv.org/abs/2210.00379.
[67]	TEWARI A, FRIED O, THIES J, et al. Advances in neural rendering[C]// SIGGRAPH '21:Special Interest Group on Computer Graphics and Interactive Techniques. New York: ACM, 2021: 1-320.
[68]	MITTAL A. Neural radiance fields: past, present, and future[EB/OL]. [2023-09-10]. https://arxiv.org/abs/2304.10050v1.
[69]	成欢, 王硕, 李孟, 等. 面向自动驾驶场景的神经辐射场综述[J]. 图学学报, 2023, 44(6): 1091-1103. DOI
	CHENG H, WANG S, LI M, et al. A review of neural radiance field for autonomous driving scene[J]. Journal of Graphics, 2023, 44(6): 1091-1103 (in Chinese). DOI
[70]	常远, 盖孟. 基于神经辐射场的视点合成算法综述[J]. 图学学报, 2021, 42(3): 376-384.
	CHANG Y, GAI M. A review on neural radiance fields based view synthesis[J]. Journal of Graphics, 2021, 42(3): 376-384 (in Chinese).
[71]	MILDENHALL B, SRINIVASAN P P, ORTIZ-CAYON R, et al. Local light field fusion: practical view synthesis with prescriptive sampling guidelines[J]. ACM Transactions on Graphics, 38(4): 29:1-29:14.
[72]	DAI A, CHANG A X, SAVVA M, et al. ScanNet: richly-annotated 3D reconstructions of indoor scenes[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2017: 2432-2443.
[73]	JENSEN R, DAHL A, VOGIATZIS G, et al. Large scale multi-view stereopsis evaluation[C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2014: 406-413.
[74]	KNAPITSCH A, PARK J, ZHOU Q Y, et al. Tanks and temples: benchmarking large-scale scene reconstruction[J]. ACM Transactions on Graphics, 36(4): 78:1-78:13.
[75]	GEIGER A, LENZ P, STILLER C, et al. Vision meets robotics: the KITTI dataset[J]. International Journal of Robotics Research, 2013, 32(11): 1231-1237.
[76]	SUN P, KRETZSCHMAR H, DOTIWALLA X, et al. Scalability in perception for autonomous driving: waymo open dataset[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2020: 2443-2451.
[77]	CAESAR H, BANKITI V, LANG A H, et al. nuScenes: a multimodal dataset for autonomous driving[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2020: 11618-11628.
[78]	LI Y X, JIANG L H, XU L N, et al. MatrixCity: a large-scale city dataset for city-scale neural rendering and beyond[C]// 2023 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2023: 3182-3192.
[79]	LU C S, YIN F K, CHEN X, et al. A large-scale outdoor multi-modal dataset and benchmark for novel view synthesis and implicit scene reconstruction[C]// 2023 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2023: 7523-7533.
[80]	ZHANG K, RIEGLER G, SNAVELY N, et al. NeRF++: analyzing and improving neural radiance fields[EB/OL]. [2023-09-10]. http://arxiv.org/abs/2010.07492.
[81]	BARRON J T, MILDENHALL B, VERBIN D, et al. Mip-NeRF 360: unbounded anti-aliased neural radiance fields[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2022: 5460-5469.
[82]	REISER C, SZELISKI R, VERBIN D, et al. MERF: memory-efficient radiance fields for real-time view synthesis in unbounded scenes[J]. ACM Transactions on Graphics, 42(4): 89.
[83]	WANG P, LIU Y, CHEN Z X, et al. F2-NeRF: fast neural radiance field training with free camera trajectories[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2023: 4150-4159.
[84]	MI Z X, XU D. Switch-NeRF: learning scene decomposition with mixture of experts for large-scale neural radiance fields[EB/OL]. [2023-09-10]. https://openreview.net/pdf?id=PQ2zoIZqvm.
[85]	XU L N, XIANGLI Y B, PENG S D, et al. Grid-guided neural radiance fields for large urban scenes[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2023: 8296-8306.
[86]	XIE Z Y, ZHANG J G, LI W Y, et al. S-NeRF: neural radiance fields for street views[EB/OL]. [2023-09-10]. http://arxiv.org/abs/2303.00749.
[87]	MARTIN-BRUALLA R, RADWAN N, SAJJADI M S M, et al. NeRF in the wild: neural radiance fields for unconstrained photo collections[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 7206-7215.
[88]	RUDNEV V, ELGHARIB M, SMITH W, et al. NeRF for outdoor scene relighting[C]// European Conference on Computer Vision. Cham: Springer, 2022: 615-631.
[89]	KARNEWAR A, RITSCHEL T, WANG O, et al. ReLU fields: the little non-linearity that could[C]// SIGGRAPH’22:Special Interest Group on Computer Graphics and Interactive Techniques Conference Proceedings. New York: ACM, 2022: 27:1-27:9.
[90]	SEHGAL S, SINGH H, AGARWAL M, et al. Data analysis using principal component analysis[C]// 2014 International Conference on Medical Imaging, m-Health and Emerging Communication Systems. New York: IEEE Press, 2014: 45-48.
[91]	KERBL B, KOPANAS G, LEIMKUEHLER T, et al. 3D Gaussian splatting for real-time radiance field rendering[J]. ACM Transactions on Graphics, 42(4): 139:1-139:14.
[92]	PARK J, JOO K, HU Z, et al. Non-local spatial propagation network for depth completion[C]// The 16th European Conference on Computer Vision. Cham: Springer, 2020: 120-136.