Abstract
Superpixels could effectively decompose an image into perceptually meaningful partitions, thus facilitating various high-level computer vision tasks. As a pre-processing tool to speed up the subsequent steps, superpixel generation is expected to be outstanding in decomposition quality and computational efficiency. In this work, we introduce the contour prior into a non-iterative clustering framework and then put forward a package of optimizations. First, contour existence between two pixels on the image plane is utilized as a scaling factor to magnify their joint color and spatial similarity. Then the inter-pixel correlation and the boundaries between different objects can be precisely described. In addition, we conduct an early assignment strategy on the conventional non-iterative clustering structure based on the smooth assumption, leading to efficient label expansion during updating and assignment. Moreover, we adopt a contour-tuned redistribution strategy to generate content-aware superpixels with the exact amount to substitute the grid sampling. Finally, these optimizations could form a synergistic framework to generate higher-quality superpixels. Extensive experiments confirm that the proposed method outperforms the baselines and achieves comparable results with state-of-the-art superpixel algorithms for several quantitative metrics.
Similar content being viewed by others
References
Ren X, Malik J (2003) Learning a classification model for segmentation. In: Proceedings of the IEEE international conference on computer vision, pp 10–17
Al-Huda Z, Peng B, Yang Y, Algburi R, Ahmad M, Khurshid F, Moghalles K (2021) Weakly supervised semantic segmentation by iteratively refining optimal segmentation with deep cues guidance. Neural Comput Appl 33:9035–9060
Loke S, MacDonald B, Parsons M, Wünsche B (2021) Accelerated superpixel image segmentation with a parallelized DBSCAN algorithm. J Real-Time Image Proc 18:2361–2376
Liu Z, Li Z, Wang R, Zong M, Ji W (2020) Spatiotemporal saliency-based multi-stream networks with attention-aware LSTM for action recognition. Neural Comput Appl 32:14593–14602
Liang Y, Wang M, Guo Y, Zhegn W (2020) On large appearance change in visual tracking. Neural Comput Appl 32:6089–6109
Wang P, Zhang J, Zhu H (2021) Fire detection in video surveillance using superpixel-based region proposal and ESE-ShuffleNet. Multimed Tools Appl 2021:1–28
Ban Z, Liu J, Cao L (2018) Superpixel segmentation using Gaussian mixture model. IEEE Trans Image Process 27(8):4105–4117
Stutz D, Hermans A, Leibe B (2018) Superpixels: An evaluation of the state-of-the-art. Comput Vis Image Underst 166:1–27
Chen J, Li Z, Huang B (2017) Linear spectral clustering superpixel. IEEE Trans Image Process 26(7):3317–3330
Wang P, Zeng G, Gan R, Wang J, Zha H (2013) Structure-sensitive superpixels via geodesic distance. Int J Comput Vis 103(1):1–21
Rubio A, Yu L, Simo-Serra E, Moreno-Noguer F (2016) BASS: boundary-aware superpixel segmentation. In: Proceedings of the IEEE international conference on pattern recognition, pp 2824–2829
Liu Y, Yu C, Yu M, He Y (2016) Manifold SLIC: a fast method to compute content-sensitive superpixels. In: Proceedings of the computer vision and pattern recognition, pp 651–659
Achanta R, Shaji A, Smith K, Lucchi A, Fua P, Susstrunk S (2012) SLIC superpixels compared to state-of-the-art superpixel methods. IEEE Trans Pattern Anal Mach Intell 34(11):2274–2282
Achanta R, Susstrunk S (2017) Superpixels and polygons using simple non-iterative clustering. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4895–4904
Giraud R, Ta V, Papadakis N (2018) Robust superpixels using color and contour features along linear path. Comput Vis Image Underst 170:1–13
Gong J, Liao N, Li C, Ma X, He W, Guo B (2021) Superpixel segmentation via contour optimized non-iterative clustering. In: Proceedings of the international conference on neural computing for advanced applications, pp 645–658
Wang M, Liu X, Gao Y, Ma X, Soomro N (2017) Superpixel segmentation: a benchmark. Signal Process Image Commun 56:28–39
Van den Bergh M, Boix X, Roig G, Van Gool L (2015) SEEDS: Superpixels extracted via energy-driven sampling. Int J Comput Vis 111(3):298–314
Liu M, Tuzel O, Ramalingam S, Chellappa R (2011) Entropy rate superpixel segmentation, In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2097–2104
Grady L (2006) Random walks for image segmentation. IEEE Trans Pattern Anal Mach Intell 28(11):1768–1783
Shen J, Hao X, Liang Z, Liu Y, Wang W, Shao L (2016) Real-time superpixel segmentation by DBSCAN clustering algorithm. IEEE Trans Image Process 25(12):5933–5942
Huang C, Wang W, Lin S, Lin Y (2016) USEQ: Ultra-fast superpixel extraction via quantization. In: Proceedings of the IEEE international conference on computer vision, pp 1965–1970
Shen J, Du Y, Wang W, Li X (2014) Lazy random walks for superpixel segmentation. IEEE Trans Image Process 23(4):1451–1462
Fu H, Cao X, Tang D, Han Y, Xu D (2014) Regularity preserved superpixels and supervoxels. IEEE Trans Multimed 16(4):1165–1175
Zhao J, Hou Q, Ren B, Cheng M, Rosin P (2018) FLIC: fast linear iterative clustering with active search. In: Proceedings of the AAAI conference on artificial intelligence, pp 7574–7581
Neubert P, Protzel P (2014) Compact watershed and preemptive SLIC: on improving trade-offs of superpixel segmentation algorithms. In: Proceedings of the IEEE international conference on pattern recognition, pp 996–1001
Hu Z, Zou Q, Li Q (2015) Watershed superpixel. In: Proceedings of the IEEE international conference on image processing, pp 349–353
Meyer F (1992) Color image segmentation. In: Proceedings of the IEEE international conference on image processing, pp 303–306
Ban Z, Liu J, Fouriaux J (2018) GLSC: LSC superpixels at over 130 FPS. J Real-Time Image Process 14(3):605–616
Liu Y, Yu M, Li B, He Y (2018) Intrinsic manifold SLIC: a simple and efficient method for computing content-sensitive superpixels. IEEE Trans Pattern Anal Mach Intell 40(3):653–666
Ye Z, Yi R, Yu M, Liu Y, He Y (2019) Fast computation of content-sensitive superpixels and supervoxels using q-distances. In: Proceedings of the IEEE international conference on computer vision, pp 3770–3779
Dollár P, Zitnick CL (2013) Structured forests for fast edge detection. In: Proceedings of the IEEE international conference on computer vision, pp 1841–1848
Choi K, Oh K (2016) Subsampling-based acceleration of simple linear iterative clustering for superpixel segmentation. Comput Vis Image Underst 146:1–8
Kang X, Zhu L, Ming A (2020) Dynamic random walk for superpixel segmentation. IEEE Trans Image Process 29:3871–3884
Arbelaez P, Maire M, Fowlkes C, Malik J (2011) Contour detection and hierarchical image segmentation. IEEE Trans Pattern Anal Mach Intell 33(5):898–916
Acknowledgements
This research is supported financially by the National Natural Science Foundation of China (62171341, 51805398), Natural Science Basic Research Program of Shaanxi Province of China (2020JM-196) and Fundamental Research Funds for the Central Universities (JB211303).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
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
Li, C., He, W., Liao, N. et al. Superpixels with contour adherence via label expansion for image decomposition. Neural Comput & Applic 34, 16223–16237 (2022). https://doi.org/10.1007/s00521-022-07315-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-022-07315-0