Abstract
Video stabilization is one of the most interesting areas of research that can be implemented in consumer camera technology. Various works have been proposed to overcome this problem. However, there are many practical difficulties in deploying them in real time. Many frames are taken as an input to generate a stabilized output, which leads to high computation of the image sensing hardware. A novel approach to solve these challenges is proposed in this paper using the Hybrid Deep Neural Network Model (HDNNM) that uses gyroscopic sensor data and the optical flow analysis. This network is based on the deep unsupervised learning techniques, which require less computation and low buffer memory. The workflow of the proposed model is based on the integration of the optical flow with the joint motion representation to maintain a correspondence between the frames and the poses of scene objects. The LSTM module in the network wraps the grid and stabilizes the video frames. The relative motion along with the multi-stage training strategies was adopted to make the model unsupervised. The HDNNM is validated using the ablative dataset collected from various scenes, and the results of the model surpass the existing model’s performance metrics. To the best of the authors’ knowledge, this is the first unsupervised approach to overcome the challenges in video stabilization.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Guilluy W, Oudre L, Beghdadi A (2021) Video stabilization: overview, challenges and perspectives. Signal Process Image Commun 90:116015. https://doi.org/10.1016/j.image.2020.116015
Greeta S, Raajan NR, Meenu M (2019) Video stabilization. In: 2019 innovations in power and advanced computing technologies (i-PACT), pp 1–4. https://doi.org/10.1109/iPACT44901.2019.8959991
Lee YG (2022) Low memory access video stabilization for low-cost camera SoC. Sensors (Basel). 22(6):2341. https://doi.org/10.3390/s22062341. PMID: 35336512; PMCID: PMC8949514
Survey on digital video stabilization: concepts, methods, and challenges. ACM Comput Surv (2023). https://dl.acm.org/doi/10.1145/3494525
Yu J, Ramamoorthi R (2019) Robust video stabilization by optimization in CNN weight space. In: CVPR
Bross B, Wang Y-K, Ye Y, Liu S, Chen J, Sullivan GJ, Ohm J-R (2021) Overview of the versatile video coding (VVC) standard and its applications. IEEE Trans Circuits Syst Video Technol 31:3736–3764
Lee YG (2019) Fast global motion estimation on single instruction multiple data processors for real-time devices. Electron Imaging 58:113105
Guo H, Liu S, Zhu S, Shen HT, Zeng B (2018) View-consistent MeshFlow for stereo-scopic video stabilization. IEEE Trans Comput Imaging 4(4):573–584
Raut S et al (2019) Real-time high-resolution video stabilization using high-frame-rate jitter sensing. ROBOMECH J (SpringerOpen). https://doi.org/10.1186/s40648-019-0144-z
Nie Y, Su T, Zhang Z, Sun H, Li G (2018) Dynamic video stitching via shakiness removing. IEEE Trans Image Process 27(1):164–178
Sen-Zhe X, Jun H, Wang M, Tai-Jiang M, Shi-Min H (2018) Deep video stabilization using adversarial networks. Comput Graph Forum 37:267–276
Wang M, Yang G-Y, Lin J-K, Zhang S-H, Shamir A, Lu S-P, Hu S-M (2019) Deep online video stabilization with multi-grid warping transformation learning. IEEE Trans Image Process
Wang Z, Zhang L, Huang H (2018) High-quality real-time video stabilization using trajectory smoothing and mesh-based warping. IEEE Access 6:25157–25166
Raptis GE, Fidas C, Avouris N (2018) Effects of mixed-reality on players behaviour and immersion in a cultural tourism game: a cognitive processing perspective. Int J Human Comput Stud 114:69–79
Kamranian Z, Sadeghian H, Nilchi ARN, Mehran M (2021) Fast, yet robust end-to-end camera pose estimation for robotic applications. Appl Intell 51(6):3581–3599
Walter H, Li R, Munafo J, Curry C, Peterson N, Stoffregen T (2019) APAL coupling study
Jia C, Evans BL (2012) Probabilistic 3-D motion estimation for rolling shutter video rectification from visual and inertial measurements. In: Proceedings of the IEEE international workshop on multimedia signal processing (MMSP), pp 203–208
Grundmann M, Kwatra V, Essa I (2011) Auto-directed video stabilization with robust L1 optimal camera paths. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), pp 225–232
Aguilar WG, Angulo C (2016) Real-time model-based video stabilization for microaerial vehicles. Neural Process Lett 43(2):459–477
Tareen SAK, Saleem Z (2018) A comparative analysis of SIFT SURF KAZE AKAZE ORB and BRISK. In: 2018 international conference on computing mathematics and engineering technologies (iCoMET), pp 1–10
Noble FK (2016) Comparison of opencv’s feature detectors and feature matchers. In: 2016 23rd international conference on mechatronics and machine vision in practice (M2VIP), pp 1–6
Casser V, Pirk S, Mahjourian R, Angelova A (2019) Depth prediction without the sensors: leveraging structure for unsupervised learning from monocular videos. AAAI 33:8001–8008
Gordon A, Li H, Jonschkowski R, Angelova A (2019) Depth from videos in the wild: unsupervised monocular depth learning from unknown cameras. In: ICCV, pp 8977–8986
Wang M, Yang G-Y, Lin J-K, Zhang S-H, Shamir A, Lu S-P, Hu S-M (2018) Deep online video stabilization with multi-grid warping transformation learning. In: IEEE TIP
Guervós E, Ruiz JJ, Pérez P, Muñoz JA, Díaz C, García N (2019) Using 360 VR video to improve the learning experience in veterinary medicine university degree. In: Electronic imaging. Society for imaging science and technology, pp 217-1–217-7
Fang M, Li H, Si S (2018) A video stabilization algorithm based on affine SIFT. In: 2018 international conference on computing, mathematics and engineering technologies (iCoMET), Sukkur, pp 1–4
Minh T, Hong M (2018) Video stabilization using feature-based classification. In: 2018 IEEE international conference on consumer electronics-Asia (ICCE-Asia), Jeju, 2018, pp 206–212
Huang TS, Netravali AN (2002) Motion and structure from feature correspondences: a review. In: Advances in image processing and understanding: a Festschrift for Thomas S Huang, pp 331–347
Luchetti A, Tomasin P, Fornaser A, Tallarico P, Bosetti P, De Cecco M (2019) The human being at the center of smart factories thanks to augmented reality. In: 2019 IEEE 5th international forum on research and technology for society and industry (RTSI), pp 51–56
Milgram P, Kishino F (1994) A taxonomy of mixed reality visual displays. IEICE Trans Inf Syst 77:1321–1329
Butaslac IIII, Luchetti A, Parolin E, Fujimoto Y, Kanbara M, De Cecco M, Kato H (2020) The feasibility of augmented reality as a support tool for motor rehabilitation. In: International conference on augmented reality, virtual reality and computer graphics. Springer, pp 165–173
Litleskare S, Calogiuri G (2019) Camera stabilization in 360\(^{\circ }\) videos and its impact on cyber sickness, environmental perceptions, and psychophysiological responses to a simulated nature walk: a singleblinded randomized trial. Front Psychol 10:2436
Zhang G, Qin X, Hua W, Wong T-T, Heng P-A, Bao H (2007) Robust metric reconstruction from challenging video sequences. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), pp 1–8
Yan C, Xie H, Yang D, Yin J, Zhang Y, Dai Q (2018) Supervised hash coding with deep neural network for environment perception of intelligent vehicles. IEEE Trans Intell Transp Syst 19(1):284–295
Dong J, Liu H (2017) Video stabilization for strict real-time applications. IEEE Trans Circuits Syst Video Technol 27(4):716–724
Kejriwal L, Singh I (2016) A hybrid filtering approach of digital video stabilization for UAV using Kalman and low pass filter. Procedia Comput Sci 93:359–366
Bian J, Li Z, Wang N, Zhan H, Shen C, Cheng M-M et al (2019) Unsupervised scale-consistent depth and egomotion learning from monocular video. NeurIPS, pp 35–45
Choi J, Kweon IS (2019) Deep iterative frame interpolation for full-frame video stabilization. In: SIGGRAPH Asia
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: CVPR, pp 770–778
Liu S, Li M, Zhu S, Zeng B (2017) Codingflow: enable video coding for video stabilization. IEEE TIP 26(7):3291–3302
Liu S, Tan P, Yuan L, Sun J, Zeng B (2016) Meshflow: minimum latency online video stabilization. In: ECCV, pp 800–815
Luo X, Huang J-B, Szeliski R, Matzen K, Kopf J (2020) Consistent video depth estimation. ACM TOG 39(4)
Ranjan A, Jampani V, Balles L, Kim K, Sun D, Wulff J et al (2019) Competitive collaboration: joint unsupervised learning of depth camera motion optical flow and motion segmentation. In: CVPR 2019-June, no. 1, pp 12232–12241
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Kasyap, V.L.V.S.K.B., Sumathi, Adhikari, A., Bhagavan, V.S. (2023). DIVS: A Real-Time Video Stabilization Framework for Consumer Camera. In: Borah, S., Gandhi, T.K., Piuri, V. (eds) Advanced Computational and Communication Paradigms . ICACCP 2023. Lecture Notes in Networks and Systems, vol 535. Springer, Singapore. https://doi.org/10.1007/978-981-99-4284-8_28
Download citation
DOI: https://doi.org/10.1007/978-981-99-4284-8_28
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4283-1
Online ISBN: 978-981-99-4284-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)