Abstract
Estimating depth from a single image presents a formidable challenge due to the inherently ill-posed and ambiguous nature of deriving depth information from a 3D scene. Prior approaches to monocular depth estimation have mainly relied on Convolutional Neural Networks (CNNs) or Vision Transformers (ViTs) as the primary feature extraction methods. However, striking a balance between speed and accuracy for real-time tasks has proven to be a formidable hurdle with these methods. In this study, we proposed a new model called EMTNet, which extracts feature information from images at both local and global scales by combining CNN and ViT. To reduce the number of parameters, EMTNet introduces the mobile transformer block (MTB), which reuses parameters from self-attention. High-resolution depth maps are generated by fusing multi-scale features in the decoder. Through comprehensive validation on the NYU Depth V2 and KITTI datasets, the results demonstrate that EMTNet outperforms previous real-time monocular depth estimation models based on CNNs and hybrid architecture. In addition, we have done the corresponding generalizability tests and ablation experiments to verify our conjectures. The depth map output from EMTNet exhibits intricate details and attains a real-time frame rate of 32 FPS, achieving a harmonious balance between real-time and accuracy.
References
Diaz, C., Walker, M., Szafir, D. A., and Szafir, D. (2017) Designing for depth perceptions in augmented reality. In: 2017 IEEE international symposium on mixed and augmented reality (ISMAR), pages 111–122. IEEE
Kusupati, U., Cheng, S., Chen, R., and Su, H. (2020) Normal assisted stereo depth estimation. In: Proceedings of the IEEE/cvf conference on computer vision and pattern recognition, pp 2189–2199
Mancini M, Costante G, Valigi P, Ciarfuglia TA (2016) Fast robust monocular depth estimation for obstacle detection with fully convolutional networks. In: 2016 IEEE/rsj international conference on intelligent robots and systems (IROS), pp 4296–4303. IEEE
Mur-Artal R, Montiel JMM, Tardós JD (2015) Orb-slam: a versatile and accurate monocular slam system. IEEE Trans Rob 31(5):1147–1163
Ha H, Im S, Park J, Jeon H-G, Kweon IS (2016) High-quality depth from uncalibrated small motion clip. In: Proceedings of the IEEE conference on computer vision and pattern Recognition, pp 5413–5421
Kong N, Black MJ (2015) Intrinsic depth: improving depth transfer with intrinsic images. In: Proceedings of the IEEE international conference on computer vision, pp 3514–3522
Karsch K, Liu C, Kang SB (2016) Depth transfer: depth extraction from videos using nonparametric sampling. In: dense image correspondences for computer vision, pp 173–205. Springer
Rajagopalan AN, Chaudhuri S, Mudenagudi U (2004) Depth estimation and image restoration using defocused stereo pairs. IEEE Trans Pattern Anal Mach Intell 26(11):1521–1525
Eigen D, Fergus R (2015) Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture. In: Proceedings of the IEEE international conference on computer vision, pp 2650–2658
Liu F, Shen C, Lin G (2015) Deep convolutional neural fields for depth estimation from a single image. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5162–5170
Porzi L, Bulo SR, Penate-Sanchez A, Ricci E, Moreno-Noguer F (2016) Learning depth-aware deep representations for robotic perception. IEEE Robotics and Autom Lett 2(2):468–475
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Łukasz, Polosukhin I (2017) Attention is all you need. Advances in neural information processing systems, pp 30
Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, Dehghani M, Minderer M, Heigold G, Gelly S, et al (2020) An image is worth 16x16 words: transformers for image recognition at scale. arXiv preprint arXiv:2010.11929
Bhat SF, Alhashim I, Wonka P (2021) Adabins: Depth estimation using adaptive bins. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 4009–4018
Li Z, Wang X, Liu X, Jiang J (2022) Binsformer: revisiting adaptive bins for monocular depth estimation. arXiv preprint arXiv:2204.00987
Zhao C, Zhang Y, Poggi M, Tosi F, Guo X, Zhu Z, Huang G, Tang Y, Mattoccia S (2022) Monovit: self-supervised monocular depth estimation with a vision transformer. arXiv preprint arXiv:2208.03543
Bae J, Moon S, Im S (2022) Deep digging into the generalization of self-supervised monocular depth estimation. arXiv preprint arXiv:2205.11083
Li Z, Chen Z, Liu X, Jiang J (2022) Depthformer: exploiting long-range correlation and local information for accurate monocular depth estimation. arXiv preprint arXiv:2203.14211
Shu C, Chen Z, Chen L, Ma K, Wang M, Ren H (2022) Sidert: A real-time pure transformer architecture for single image depth estimation. arXiv preprint arXiv:2204.13892
Ma H, Xia X, Wang X, Xiao X, Li J, Zheng M (2022) MoCoViT: mobile convolutional vision transformer
Ranftl R, Bochkovskiy A, Koltun V (2021) Vision transformers for dense prediction. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 12179–12188
Silberman N, Hoiem D, Kohli P, Fergus R (2012) Indoor segmentation and support inference from rgbd images. In: European conference on computer vision, pp 746–760. Springer
Geiger A, Lenz P, Stiller C, Urtasun R (2013) Vision meets robotics: the kitti dataset. Int J Robotics Res 32(11):1231–1237
Saxena A, Chung S, Ng A (2005) Learning depth from single monocular images. Advances in neural information processing systems, 18
Karsch K, Liu C, Kang SB (2019) Depth extraction from video using non-parametric sampling. arXiv preprint arXiv:2002.04479
Konrad J, Wang M, Ishwar P (2012) 2d-to-3d image conversion by learning depth from examples. In: 2012 IEEE Computer society conference on computer vision and pattern recognition workshops, pp 16–22. IEEE
Karsch K, Liu C, Kang SB (2014) Depth transfer: depth extraction from video using non-parametric sampling. IEEE Trans Pattern Anal Mach Intell 36(11):2144–2158
Liu M, Salzmann M, He X (2014) Discrete-continuous depth estimation from a single image. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 716–723
Eigen D, Puhrsch C, Fergus R (2014) Depth map prediction from a single image using a multi-scale deep network. Advances in neural information processing systems, 27
Laina I, Rupprecht C, Belagiannis V, Tombari F, Navab N (2016) Deeper depth prediction with fully convolutional residual networks. In: 2016 Fourth international conference on 3D vision (3DV), pp 239–248. IEEE
Fu H, Gong M, Wang C, Batmanghelich K, Tao D (2018) Deep ordinal regression network for monocular depth estimation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2002–2011
Liu F, Shen C, Lin G, Reid I (2015) Learning depth from single monocular images using deep convolutional neural fields. IEEE Trans Pattern Anal Mach Intell 38(10):2024–2039
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. 2016 IEEE conference on computer vision and pattern recognition (CVPR), pp 770–778
Qi X, Liao R, Liu Z, Urtasun R, Jia J (2018) Geonet: geometric neural network for joint depth and surface normal estimation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 283–291
Touvron H, Cord M, Douze M, Massa F, Sablayrolles A, Jégou H (2021) Training data-efficient image transformers & distillation through attention. In: International conference on machine learning, pp 10347–10357. PMLR
Wang W, Xie E, Li X, Fan D-P, Song K, Liang D, Lu T, Luo P, Shao L (2021) Pyramid vision transformer: a versatile backbone for dense prediction without convolutions. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 568–578
Han K, Xiao A, Enhua W, Guo J, Chunjing X, Wang Y (2021) Transformer in transformer. Adv Neural Inf Process Syst 34:15908–15919
Xu W, Xu Y, Chang T, Tu Z (2021) Co-scale conv-attentional image transformers. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 9981–9990
Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, Lin S, Guo B (2021) Swin transformer: hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 10012–10022
Dalmaz O, Yurt M, Çukur T (2021) Resvit: residual vision transformers for multimodal medical image synthesis. IEEE Trans Med Imaging 41:2598–2614
Mehta S, Rastegari M (2021) Mobilevit: Light-weight, general-purpose, and mobile-friendly vision transformer. arXiv:2110.02178
Li Z, Li Y, Li Q, Zhang Y, Wang P, Guo D, Lu L, Jin D, Hong Q (2022) Lvit: Language meets vision transformer in medical image segmentation. arXiv:2206.14718
Wu K, Zhang J, Peng H, Liu M, Xiao B, Fu J, Yuan L (2022) Tinyvit: fast pretraining distillation for small vision transformers. arXiv:2207.10666
Dai Z, Liu H, Le QV, Tan M (2021) Coatnet: marrying convolution and attention for all data sizes. Adv Neural Inf Process Syst 34:3965–3977
Srinivas A, Lin T-Y, Parmar N, Shlens J, Abbeel P, Vaswani A (2021) Bottleneck transformers for visual recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 16519–16529
d’Ascoli S, Touvron H, Leavitt ML, Morcos AS, Biroli G, Sagun L (2021) Convit: improving vision transformers with soft convolutional inductive biases. In: International Conference on Machine Learning, pp 2286–2296. PMLR
Wu H, Xiao B, Codella N, Liu M, Dai X, Yuan L, Zhang L (2021) Cvt: introducing convolutions to vision transformers. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 22–31
Graham B, El-Nouby A, Touvron H, Stock P, Joulin A, Jégou H, Douze M (2021) Levit: a vision transformer in convnet’s clothing for faster inference. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 12259–12269
LeCun Y, Boser B, Denker J, Henderson D, Howard R, Hubbard W, Jackel L (1989) Handwritten digit recognition with a back-propagation network. Advances in neural information processing systems, 2
Han K, Wang Y, Tian Q, Guo J, Xu C, Xu C (2020) Ghostnet: more features from cheap operations. In: 2020 IEEE/CVF conference on computer vision and pattern recognition (CVPR), pp 1577–1586
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 7132–7141
Lee JH, Han M-K, Ko DW, Suh IH (2019) From big to small: multi-scale local planar guidance for monocular depth estimation. arXiv preprint arXiv:1907.10326
Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A (2010) The pascal visual object classes (voc) challenge. Int J Comput Vision 88(2):303–338
Smith LN, Nicholay T (2019) Super-convergence: very fast training of neural networks using large learning rates. In: Tien Pham (ed) Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications. International Society for Optics and Photonics, SPIE, vol 11006, pp 1100612
Nekrasov V, Dharmasiri T, Spek A, Drummond T, Shen C, Reid ID (2018) Real-time joint semantic segmentation and depth estimation using asymmetric annotations. In: 2019 international conference on robotics and automation (ICRA), pp 7101–7107
Wofk D, Ma F, Yang T-J, Karaman S, Sze V (2019) Fastdepth: fast monocular depth estimation on embedded systems. In: 2019 international conference on robotics and automation (ICRA), IEEE, pp 6101–6108.
Spek A, Dharmasiri T, Drummond T (2018) Cream: condensed real-time models for depth prediction using convolutional neural networks. 2018 IEEE/RSJ international conference on intelligent robots and systems (IROS), pp 540–547
Wang L, Famouri M, Wong A (2020) Depthnet nano: a highly compact self-normalizing neural network for monocular depth estimation. arXiv:2004.08008
Ma F, Karaman S (2017) Sparse-to-dense: depth prediction from sparse depth samples and a single image. 2018 IEEE international conference on robotics and automation (ICRA), pp 1–8
Yucel MK, Dimaridou V, Drosou A, Saà-Garriga A (2021) Real-time monocular depth estimation with sparse supervision on mobile. 2021 IEEE/CVF conference on computer vision and pattern recognition workshops (CVPRW), pp 2428–2437
An S, Zhou F, Yang M, Zhu H, Fu C, Tsintotas KA (2021) Real-time monocular human depth estimation and segmentation on embedded systems. 2021 IEEE/RSJ international conference on intelligent robots and systems (IROS), pp 55–62
Atapour-Abarghouei A, Breckon TP (2018) Real-time monocular depth estimation using synthetic data with domain adaptation via image style transfer. In: 2018 IEEE/CVF conference on computer vision and pattern recognition, pp 2800–2810
Klingner M, Termöhlen J-A, Mikolajczyk J, Fingscheidt T (2020) Self-supervised monocular depth estimation: solving the dynamic object problem by semantic guidance. In: european conference on computer vision
Liu J, Li Q, Cao R, Tang W, Qiu G (2020) MiniNet: an extremely lightweight convolutional neural network for real-time unsupervised monocular depth estimation. ISPRS J Photogrammetry Remote Sens 166:255–267
Bae J-H, Moon S, Im S (2022) Deep digging into the generalization of self-supervised monocular depth estimation. Proceedings of the AAAI conference on artificial intelligence
Zhang N, Nex F, Vosselman G, Kerle N (2022) Lite-mono: a lightweight cnn and transformer architecture for self-supervised monocular depth estimation. arXiv:2211.13202
Varma A, Chawla H, Zonooz B, Arani E (2022) Transformers in self-supervised monocular depth estimation with unknown camera intrinsics. arXiv:2202.03131
Bhat SF, Alhashim I, Wonka P (2021) Adabins: depth estimation using adaptive bins. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 4009–4018
Howard AG, Zhu M, Chen B, Kalenichenko D, Wang W, Weyand T, Andreetto M, Adam H (2017) Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yan, L., Yu, F. & Dong, C. EMTNet: efficient mobile transformer network for real-time monocular depth estimation. Pattern Anal Applic 26, 1833–1846 (2023). https://doi.org/10.1007/s10044-023-01205-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-023-01205-4