Junqi Shi

Junqi Shi

Auto-generated profile, based on publicly available information

Junqi Shi

Nanjing University

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
9 October 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
JSJunqi Shi
Junqi Shi
MLMing Lu
Ming Lu
ZMZhan Ma
Zhan Ma
Quantizing a floating-point neural network to its fixed-point representation is crucial for Learned Image Compression (LIC) because it improves decoding consistency for interoperability and reduces space-time complexity for implementation. Existing solutions often have to retrain the network for model quantization, which is time-consuming and impractical to some extent. This work suggests using Post-Training Quantization (PTQ) to process pretrained, off-the-shelf LIC models. We theoretically prove that minimizing quantization-induced mean square error (MSE) of model parameters (e.g., weight, bias, and activation) in PTQ is sub-optimal for compression tasks and thus develop a novel Rate-Distortion (R-D) Optimized PTQ (RDO-PTQ) to best retain the compression performance. Given a LIC model, RDO-PTQ layer-wisely determines the quantization parameters to transform the original floating-point parameters in 32-bit precision (FP32) to fixed-point ones at 8-bit precision (INT8), for which a tiny calibration image set is compressed in optimization to minimize R-D loss. Experiments reveal the outstanding efficiency of the proposed method on different LICs, showing the closest coding performance to their floating-point counterparts. Our method is a lightweight and plug-and-play approach without retraining model parameters but just adjusting quantization parameters, which is attractive to practitioners. Such an RDO-PTQ is a task-oriented PTQ scheme, which is then extended to quantize popular super-resolution and image classification models with negligible performance loss, further evidencing the generalization of our methodology. Related materials will be released at https://njuvision.github.io/RDO-PTQ.

28 October 2024
Verified
Conference Paper
Open access
JSJunqi Shi
Junqi Shi
MJMingyi Jiang
Mingyi Jiang
MLMing Lu
Ming Lu
TCTong Chen
Tong Chen
XCXun Cao
Xun Cao
ZMZhan Ma
Zhan Ma
This paper introduces HINER, a novel neural representation for compressing HSI and ensuring high-quality downstream tasks on compressed HSI. HINER fully exploits inter-spectral correlations by explicitly encoding of spectral wavelengths and achieves a compact representation of the input HSI sample through joint optimization with a learnable decoder. By additionally incorporating the Content Angle Mapper with the L1 loss, we can supervise the global and local information within each spectral band, thereby enhancing the overall reconstruction quality. For downstream classification on compressed HSI, we theoretically demonstrate the task accuracy is not only related to the classification loss but also to the reconstruction fidelity through a first-order expansion of the accuracy degradation, and accordingly adapt the reconstruction by introducing Adaptive Spectral Weighting. Owing to the monotonic mapping of HINER between wavelengths and spectral bands, we propose Implicit Spectral Interpolation for data augmentation by adding random variables to input wavelengths during classification model training. Experimental results on various HSI datasets demonstrate the superior compression performance of our HINER compared to the existing learned methods and also the traditional codecs. Our model is lightweight and computationally efficient, which maintains high accuracy for downstream classification task even on decoded HSIs at high compression ratios. Our materials will be released at https://github.com/Eric-qi/HINER.

14 November 2025 in IEEE Journal of Selected Topics in Signal Processing
Verified
Journal Article
Closed access
JSJunqi Shi
Junqi Shi
QZQirui Zhang
Qirui Zhang
MLMing Lu
Ming Lu
ZMZhan Ma
Zhan Ma
Hyperspectral image (HSI) signals are high-dimensional, continuous, and sensor-variant, posing significant challenges for compression and restoration-especially under the scarcity of labeled data and the limitations of discretized, supervised models. In this work, we propose HINER++, a unified, self-supervised framework based on implicit neural representations (INRs) for continuous hyperspectral signal modeling. HINER++ formulates each HSI sample as a wavelength-to-intensity mapping, parameterized by a compact neural network. Thus, HSI signal modeling becomes equivalent to a continuous neural function regression problem, where compression is achieved by encoding the parameters of the neural function. Building on the principles of Deep Image Prior (DIP), we further reinterpret the inductive bias of the INR architecture as a natural prior for HSI restoration, enabling tasks such as denoising, inpainting, and super-resolution without labeled supervision. Finally, we investigate the impact of lossy INR-based compression on downstream perception tasks, using classification as a proxy, and demonstrate that such implicit modeling helps preserve task performance under extreme compression via proposed task-aware components-Adaptive Spectral Weighting (ASW) and Implicit Spectral Interpolation (ISI). Extensive experiments on real-world HSI benchmarks demonstrate that HINER++ achieves superior performance across multiple compression and restoration tasks, preserving classification accuracy under extreme compression ratios (up to 100×). This work offers a new perspective: compression is not merely data reduction, but a self-supervised process of signal disentanglement and recovery. All source code and data will be made publicly available at https://github.com/Eric-qi/HINER to support reproducible research.

1 November 2026 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
XLXingchen Li
Xingchen Li
JZJunzhe Zhang
Junzhe Zhang
JSJunqi Shi
Junqi Shi
MLMing Lu
Ming Lu
ZMZhan Ma
Zhan Ma

10 November 2022
Verified
Conference Paper
Closed access
PDF file
JSJunqi Shi
Junqi Shi
LLLei Li
Lei Li
FLFuxiang Liu
Fuxiang Liu
CXChunfeng Xu
Chunfeng Xu
With the development of remote sensing technology, remote sensing images are developing towards higher resolution and larger data volume, and it is important to identify objects from remote sensing images quickly and accurately. Due to the significant differences between remote sensing images and natural images, the direct application of existing algorithms to remote sensing images is not ideal. Therefore, we take the YOLOv5 object detection algorithm as base, and have completed the following tasks: First, we present an adaptive image cutting data preprocessing method, which fills or cuts images into uniform size to cope with the resolution differences of remote sensing images. Second, we use the Mosaic data enhancement method to improve the algorithm's effect in complex backgrounds. Third, we use the Soft-NMS postprocessing algorithm to reduce the missed detection of dense objects. Furthermore, we transplant the algorithm to the hardware platform. After the above improvements, the mAP of our algorithm increases from 0.354 to 0.677 on the DOTA dataset, achieving a good object detection effect on remote sensing images; and with the help of the TensorRT, it has reached a detection speed of about 60 FPS on the NVIDIA Jetson AGX Xavier embedded hardware platform.

1 January 2026 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Verified
Journal Article
Closed access
HLHanfei Li
Hanfei Li
AKAnle Ke
Anle Ke
JSJunqi Shi
Junqi Shi
TCTong Chen
Tong Chen
ZMZhan Ma
Zhan Ma