Yifan Sun has a total of 39 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • RTPSeg: A multi-modality dataset for LiDAR point cloud semantic segmentation assisted with RGB-thermal images in autonomous driving

    by Yifan Sun, Chenguang Dai, Wenke Li, Xinpu Liu, Yongqi Sun, Ye Zhang, Weijun Guan, Yongsheng Zhang, Yulan Guo, Hanyun Wang


  • Supervised Contrastive Learning for Indoor Point Cloud Oversegmentation

    by Yifan Sun, Chenguang Dai, Wenke Li, Yongsheng Zhang, Song Ji, Anzhu Yu, Yiping Chen, Hanyun Wang

    Point cloud oversegmentation method can obtain a series of superpoints by grouping points that are semantically and geometrically consistent. The generated superpoints can be treated as the basic processing units in various downstream tasks to improve task performance and processing efficiency. However, due to the high semantic and geometric complexity of point cloud scenes, obtaining high-quality superpoints is still challenging. Aiming to generate high-quality indoor superpoints, we propose an end-to-end supervised contrastive learning framework SCL-OverSeg for indoor point cloud oversegmentation. Firstly, to solve the challenge of balancing the importance of geometric similarity and spatial proximity constraint between points and superpoints in indoor scenes, we integrate the geometric similarity and spatial proximity constraint into the supervision signal by generating the superpoint ground truth. To solve the challenge of superpoints crossing objects, we propose to utilize instance labels rather than semantic labels to generate the ideal superpoint ground truth as the object-level supervision signal. Secondly, to construct the distinguishable embedding space facilitating to the assignments of points to superpoints, we propose point-superpoint contrastive learning to compel the network to project each point to be closer to the reasonable superpoint in embedding space. Besides, with the instance labels, to improve the superpoint performance on object boundaries, we propose the object boundary contrastive learning to enhance the feature distinguishability between tough points across the object boundaries. Extensive experiments demonstrate that SCL-OverSeg can effectively improve indoor oversegmentation performance, especially on object boundaries. The relevant codes will be available on https://github.com/sssssyf/SCL-OverSeg.


  • Radar Signal Deinterleaving Based on Hidden Markov Chains and Residual Fence Networks

    by Min Xie, Jie Huang, Chuang Zhao, De-Xiu Hu, Yi-Fan Sun

    For prior-knowledge-informed scenarios, this paper proposes a radar signal deinterleaving method based on hidden Markov chains and residual fence networks (RFNs) with enhanced applicability for different pulse repetition interval (PRI) types and improved performance under PRI jitter. The proposed approach accommodates the five main PRI modulation types by modeling interleaved pulse streams as hidden Markov models (HMMs). Pulse deinterleaving is transformed into a state-sequence prediction problem using HMMs and further into a path optimization problem within RFNs. This method utilizes global information for reliable sequence separation. Experimental results indicate that the method effectively deinterleaves fixed, staggered, dwell-and-switch, sliding, and wobulated sequences under unintentional jitter, significantly improving performance under intentional jitter. The practical value and applicability of the method are demonstrated, providing a new perspective for deinterleaving. The code will be open at: https://github.com/xm980426/HMC-RFN.git .


  • Deinterleaving Method for Radar Signals With Dwell and Switch Pulse Repetition Intervals in a Prior Unknown Scenario

    by Min Xie, Chuang Zhao, Jie Huang, Yi-Fan Sun, De-Xiu Hu, Ze-Wen Wang

    Dwell and switch (D&S) pulse repetition intervals (PRIs) are vital in modern radar systems. However, the existing deinterleaving methods that do not rely on prior information are generally ineffective at handling D&S signals. Considering the characteristics of D&S PRIs, spectrum analysis and maximal clique are employed in this study to deinterleave D&S signals in a prior unknown scenario. Sequence search technology is introduced for D&S subsequences to ensure complete D&S subsequences within the search results. Furthermore, D&S subsequence identification technology based on spectrum analysis is developed. Following vector processing of the search results, D&S subsequences are distinguished from fixed sequences based on the number of spectral peaks within the principal interval. Additionally, a subsequence consolidation technique based on the maximal clique is introduced, where the sequence search result is regarded as the vertex, and the existence of an edge between corresponding vertices is determined by evaluating the complementarity and mutual exclusion of the combination. Through the construction of an undirected graph, subsequences from the same D&S sequence are consolidated using the maximal clique extraction. The simulation results demonstrate that the detection threshold can effectively identify subsequences, whereas the complementarity and mutual exclusion constraints efficiently detect associations between subsequences, mitigating false association pairs from false alarms of fixed sequences. For a pulse missing rate of 30% or less, the analysis accuracy of D&S parameters reaches 98.23%, highlighting the outstanding performance and promising application prospects of the proposed method in deinterleaving D&S sequences and decoding their modulation patterns.


  • Hyperspectral Meets Optical Flow: Spectral Flow Extraction for Hyperspectral Image Classification

    by Bing Liu, Yifan Sun, Anzhu Yu, Zhixiang Xue, Xibing Zuo

    Hyperspectral image (HSI) classification has always been recognised as a difficult task. It is therefore a research hotspot in remote sensing image processing and analysis, and a number of studies have been conducted to better extract spectral and spatial features. This study aimed to track the variation of the spectrum in hyperspectral images from a sequential data perspective to obtain more distinguishable features. Based on the characteristics of optical flow, this study introduces an optical flow technique for the extraction of spectral flow that denotes the spectral variation and implements a dense optical flow extraction method based on deep matching. Lastly, the extracted spectral flow are combined with the original spectral features and input into a commonly used support vector machine (SVM) classifier to complete the classification. Extensive classification experiments on three benchmark HSI test sets show that the classification accuracy obtained by the spectral flow extracted in this study (SpectralFlow) is higher than traditional spatial feature extraction methods, texture feature extraction methods, and the latest deep-learning-based methods. Furthermore, the proposed method can produce finer classification thematic maps, thereby demonstrating strong practical application potential.


  • Spectral–Spatial MLP-Like Network With Reciprocal Points Learning for Open-Set Hyperspectral Image Classification

    by Yifan Sun, Bing Liu, Ruirui Wang, Pengqiang Zhang, Mofan Dai

    In recent years, deep-learning-based hyperspectral image (HSI) classification methods have achieved significant development and gradually become widely applied. The existing advanced methods can achieve near-saturation performance with sufficient labels in a closed-set environment (CSE), i.e., training set and test set are all known categories of ground objects. However, the real world is usually open because of the diversity of land covers, i.e., test-set exists unknown categories that are not labeled in the training set. Therefore, the prevalent advanced CSE methods still cannot effectively and robustly handle unknown categories of ground objects in an open-set environment (OSE). Therefore, we propose a spectral-spatial MLP-like network with reciprocal points learning (SSMLP-RPL) to improve the performance of open-set HSI classification. First, a feature learning framework based on reciprocal points learning (RPL) is constructed to model the extra-category space and reduce the risk of open space. The learned feature space enables to enlarge the distance between the known and unknown categories. Besides, we further propose to utilize a learnable dynamic threshold of each known category to effectively distinguish the unknown categories and improve open performance of the model. Second, to enhance the capacity of feature learning, a spectral-spatial MLP-like network (SSMLP) is designed to capture the spectral-spatial feature merely with a series of fully-connected (FC) layers, which mainly involve SpeFC and SpaFC two modules. Among them, the SpaFC module enables to model spacial semantics, and the SpeFC module enables to model long-distance spectral dependence. Extensive experiments on three benchmark HSIs show that SSMLP-RPL has a competitive performance both in CSE and OSE and even surpasses currently advanced closed-set and open-set HSI classification methods. As an end-to-end HSI classification framework of MLP-backbone, SSMLP network can compete with the advanced works based on CNN and transformer. The code will be open at: https://github.com/sssssyf/SSMLP-RPL.


  • Exploiting Discriminative Advantage of Spectrum for Hyperspectral Image Classification: SpectralFormer Enhanced by Spectrum Motion Feature

    by Yifan Sun, Bing Liu, Xuchu Yu, Anzhu Yu, Pengqiang Zhang, Zhixiang Xue

    As for hyperspectral image (HSI), the discrepancy of contiguous spectral information should be the main basis for the identification of ground objects. Due to the difficulty of spectral sequence coding and the spectrum similarity between categories, successful deep-learning-based classification methods always attempt to capture the spatial information to improve the accuracy by convolutional neural networks (CNN) or other excellent spatial feature extractors. However, extracting spatial feature is generally accompanied by the distortion of ground objects distribution and categories boundary. To effectively represent spectral feature, the SpectralFormer based on transformer backbone can better capture the long-term dependence of spectrum, which improves the performance of spectral feature methods significantly. However, it is still unable to compete with advanced spectral-spatial feature methods. To exploit the discriminative advantage of spectrum fully, this letter introduces an efficient sparse-to-dense optical flow estimation method to track the spectrum variation in the HSI. Then, we take such variation as a spectrum motion feature to enhance the original spectrum. At last, we continue to use the SpectralFormer to encode the concatenated spectrum sequence for classification. Extensive experiments show that the SpectralFormer enhanced by spectrum motion feature (SF-SMF) significantly improves the performance of spectral feature methods, even surpassing advanced spectral-spatial feature methods. SF-SMF can avoid interference with additional spatial information to obtain exquisite whole-domain classification maps, showing its practical value. The codes will be public at https://github.com/sssssyf/SF-SMF.


  • Top Cited Papers

    Perceiving Spectral Variation: Unsupervised Spectrum Motion Feature Learning for Hyperspectral Image Classification

    by Yifan Sun, Bing Liu, Xuchu Yu, Anzhu Yu, Kuiliang Gao, Lei Ding

    In recent years, deep-learning-based hyperspectral image (HSI) classification methods have achieved significant development. The superior capability of feature extraction from these data-driven methods dramatically improves the classification performance. However, the previous methods usually require to retrain the network from scratch to obtain the capability of feature extraction adaptive for the target image when facing a new HSI to be classified, which is a time-consuming and redundant process. In this paper, we consider putting this process ahead and making the network have a robust capability of feature extraction with generalization through pre-training. Therefore, the network enables to directly extract features of the target HSI without re-training. For this purpose, we rethink the three-dimension (3D) HSI data from a perspective of spectral sequence, and we attempt to extract the spectral variation information as the spectrum motion feature. Then, we construct an unsupervised spectrum motion feature learning framework (SMF-UL), which can be pre-trained on mass unlabeled HSI data to learn the knowledge about perceiving spectral variation. Furthermore, to achieve the expansion of source data for pre-training, we develop an extendable training dataset construction method, which can integrate HSIs of different sizes, number of bands and sensors into a unified training set to utilize the rapidly growing mass unlabeled HSI data effectively. Finally, we use the trained network to directly extract the spectrum motion feature of the target HSI for classification, so the laborious re-training of the network can be avoided. Extensive experiments show that the proposed SMF-UL acquires the robust capability of feature extraction with generalization through unsupervised learning on mass unlabeled HSI data, and the classification performance of extracted spectrum motion feature is competitive to advanced in-domain and cross-domain methods, which shows its flexibility and superiority. The code of SMF-UL will be open at: https://github.com/sssssyf/SMF-UL.