Haoyang Cheng

Dr. Haoyang Cheng


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 October 2024 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
HWHaitao Wen
Haitao Wen
HQHeqian Qiu
Heqian Qiu
LWLanxiao Wang
Lanxiao Wang
HCHaoyang Cheng
Haoyang Cheng
HLHongliang Li
Hongliang Li
Catastrophic forgetting is the core problem of class incremental learning (CIL). Existing work mainly adopts memory replay, knowledge distillation, and dynamic architecture to alleviate this problem, but seldom from the aspect of parameter regularization. However, existing parameter regularization methods struggle to achieve an appropriate balance between old and new tasks. To bring it back to CIL, we first propose constrained incremental learning with less forgetting direction (LFD) to leave more plasticity for the new task under a strong stability constraint for old tasks. Specifically, the new parameters are constrained to be close to the LFD of old tasks instead of a single group of old parameters. To validate the effectiveness of this regularization, we investigate the connectivity between the old parameters and the new parameters, and additionally find that a higher accuracy interval exists along the linear connection. Therefore, we further propose a post-processing procedure to find an equilibrium point in this interval for better balance between old and new tasks. Extensive classification experiments on CIFAR-100, ImageNet-100, and ImageNet-1K show our method can significantly improve performance compared with existing CIL methods and the object detection experiments on PASCAL-VOC show its broad generality on other tasks.

3 May 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
HCHaoyang Cheng
Haoyang Cheng
HLHongliang Li
Hongliang Li
Qingbo WuQingbo Wu
Qingbo Wu
HQHeqian Qiu
Heqian Qiu
XZXiaoliang Zhang
Xiaoliang Zhang
TZTaijin Zhao
Taijin Zhao
Contrastive learning has gained great prominence recently, which achieves excellent performance by simple augmentation invariance. However, the simple contrastive pairs suffer from lacking of diversity due to the mechanical augmentation strategies. In this paper, we propose Disturbed Augmentation Invariance (DAI for abbreviation), which constructs disturbed contrastive pairs by generating appropriate disturbed views for each augmented view in the feature space to increase the diversity. In practice, we establish a multivariate normal distribution for each augmented view, whose mean is corresponding augmented view and covariance matrix is estimated from its nearest neighbors in the dataset. Then we sample random vectors from this distribution as the disturbed views to construct disturbed contrastive pairs. In order to avoid extra computational cost with the increase of disturbed contrastive pairs, we utilize an upper bound of the trivial disturbed augmentation invariance loss to construct the DAI loss. In addition, we propose Bottleneck version of Disturbed Augmentation Invariance (BDAI for abbreviation) inspired by the Information Bottleneck principle, which further refines the extracted information and learns a compact representation by additionally increasing the variance of the original contrastive pair. In order to make BDAI work effectively, we design a statistical strategy to control the balance between the amount of the information shared by all disturbed contrastive pairs and the compactness of the representation. Our approach gets a consistent improvement over the popular contrastive learning methods on a variety of downstream tasks, e.g. image classification, object detection and instance segmentation.

26 August 2024 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
MZMinjian Zhang
Minjian Zhang
HQHeqian Qiu
Heqian Qiu
LWLanxiao Wang
Lanxiao Wang
HCHaoyang Cheng
Haoyang Cheng
TZTaijin Zhao
Taijin Zhao
HLHongliang Li
Hongliang Li
Considering the arbitrary orientation of remote sensing objects, accurate angle prediction plays a crucial role in achieving precise oriented object detection (OOD) of aerial scenes. Existing transformer-based methods typically adopt an iterative refinement mechanism to update the angle prediction and perform bipartite graph matching based on the combined matching costs. However, these methods may suffer from the angle error accumulation across decoder layers and inconsistency between the L1 cost and the rotated IoU cost, thus resulting in inaccurate angle prediction. To address these problems, this paper proposes a novel transformer-based OOD method named Oriented-DINO, which comprises three important components: Error-mitigating Angle Decoupling Prediction (EADP) module, Nonlinear Angle-conversion Consistency Optimizer (NACO), and Query-driven Diversity (QD) loss. To mitigate the angle error, the EADP module decouples the angle prediction from the iterative box refinement process and employs independent branches to directly predict the angle. To address the issue of inconsistent matching, the NACO module employs a nonlinear function for angle conversion in matching cost calculation. This approach effectively alleviates the matching cost discrepancy in angle boundary case, while preserving the consistency in other instances. To avoid highly overlapped predictions triggered by similar queries, we introduce the QD loss to encourage the generation of diverse object queries, thus avoiding redundant predictions and enhancing prediction accuracy. Extensive experimental results demonstrate that our method achieves superior performance on OOD task.

24 February 2023 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
HCHaoyang Cheng
Haoyang Cheng
HLHongliang Li
Hongliang Li
HQHeqian Qiu
Heqian Qiu
Qingbo WuQingbo Wu
Qingbo Wu
XZXiaoliang Zhang
Xiaoliang Zhang
KNKing Ngi Ngan
King Ngi Ngan
Recently, contrastive learning based on augmentation invariance and instance discrimination has made great achievements, owing to its excellent ability to learn beneficial representations without any manual annotations. However, the natural similarity among instances conflicts with instance discrimination which treats each instance as a unique individual. In order to explore the natural relationship among instances and integrate it into contrastive learning, we propose a novel approach in this paper, Relationship Alignment (RA for abbreviation), which forces different augmented views of current batch instances to main a consistent relationship with other instances. In order to perform RA effectively in existing contrastive learning framework, we design an alternating optimization algorithm where the relationship exploration step and alignment step are optimized respectively. In addition, we add an equilibrium constraint for RA to avoid the degenerate solution, and introduce the expansion handler to make it approximately satisfied in practice. In order to better capture the complex relationship among instances, we additionally propose Multi-Dimensional Relationship Alignment (MDRA for abbreviation), which aims to explore the relationship from multiple dimensions. In practice, we decompose the final high-dimensional feature space into a cartesian product of several low-dimensional subspaces and perform RA in each subspace respectively. We validate the effectiveness of our approach on multiple self-supervised learning benchmarks and get consistent improvements compared with current popular contrastive learning methods. On the most commonly used ImageNet linear evaluation protocol, our RA obtains significant improvements over other methods, our MDRA gets further improvements based on RA to achieve the best performance. The source code of our approach will be released soon.


12 June 2026 in Electronics
Verified
Journal Article
Open access
PDF file
HCHaoyang Cheng
Haoyang Cheng
XLXiao Li
Xiao Li
Qi TianQi Tian
Qi Tian
WHWei Han
Wei Han
XZXiaoliang Zhang
Xiaoliang Zhang
ZYZheng Yang
Zheng Yang
Radar signal sorting (RSS) and radar emitter recognition (RER) constitute foundational yet challenging operations in electronic reconnaissance, where RSS aims to accurately segregate interleaved radar pulse streams and RER aims to recognize their originating emitters. Existing methods typically address RSS and RER as separate processes within a sequential streaming framework, which neglect the inherent interdependence and collaborative potential between them, thereby resulting in error accumulation and performance bottleneck. In this paper, we redefine the radar signal sorting and recognition (RSSR) problem from an integrated modeling perspective, decomposing it into three sub-problems, i.e., signal pattern detection, signal pattern extraction, and detection result integration. In order to effectively solve these problems, we propose a novel Unified Framework inspired by Object Detection (UFiOD). Firstly, an end-to-end neural network is constructed to simultaneously optimize the regression of signal temporal occurrence regions and the recognition of signal categories. Then, a template matching algorithm is designed to extract corresponding pulses from the regions based on the signal categories. Finally, an integration algorithm based on temporal correlation and direction of arrival (DOA) fuses the detection results to generate object-level sorting and recognition conclusions. We extensively validate the effectiveness of the proposed method on simulation datasets. It demonstrates robust performance under various interleaving scenarios, including the interleaving of homogeneous radar emitters. Notably, it exhibits impressive capability for handling unknown signals, further highlighting its practical utility.