Computer Vision and Pattern Recognition Based on Machine Learning

Dear Colleagues,

Computer vision plays a pivotal role in modern intelligent systems, with widespread applications in medical diagnostics, industrial automation, remote sensing, autonomous driving, and beyond. Tasks such as image recognition, object detection, semantic segmentation, and 3D pose estimation are fundamental to these domains. However, the increasing complexity of vision models and the growing demand for real-time, resource-efficient solutions necessitate advancements in data-efficient and model-efficient learning.

This Special Issue focuses on cutting-edge research in efficient computer vision algorithms, covering both data efficiency and model efficiency. In data-efficient learning, the explored techniques reduce reliance on large-scale labelled datasets, including few-shot learning, transfer learning, domain adaptation, and self-supervised pre-training for downstream task adaptation. In model-efficient learning, the emphasis lies on lightweight architectures, neural architecture search (NAS), model compression (e.g., pruning, quantization, distillation), and the efficient design of convolutional neural networks (CNNs), vision transformers (ViTs), and hybrid architectures.

 This Special Issue welcomes original research and reviews addressing efficiency challenges in computer vision. Other potential applications may include, but are not limited to, the following:

 • Medical imaging (e.g., low-data lesion detection);

• Industrial inspection (e.g., defect recognition with limited samples);

• Autonomous systems (e.g., real-time object tracking);

• Remote sensing (e.g., efficient land-cover segmentation). 

In summary, we welcome studies on the following topics:

1. Data-Efficient Learning

      - Few-shot/zero-shot learning for vision tasks;  

      - Transfer learning and domain adaptation;  

      -Self-supervised and weakly supervised learning;  

      - Active learning and annotation-efficient methods.

2. Model-Efficient Design  

      -Efficient CNN and transformer architectures;  

      - Neural architecture search (NAS) for efficient models;  

      - Model compression (pruning, quantization, knowledge distillation);  

      - Dynamic or adaptive inference for computational savings.

3. Efficient Vision Tasks

      -Real-time object detection and segmentation;

      - Efficient depth estimation and 3D reconstruction

      - Low-latency video analysis (e.g., action recognition);

      - Energy-efficient deployment on edge devices.

4. Applications and Case Studies

   - Efficient vision systems for healthcare, robotics, or agriculture;

   - Benchmarks and datasets for evaluating efficiency;

   - Hardware-aware algorithm design (e.g., for mobile/embedded devices).

Dr. Tao Wang
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• computer vision
• architecture design
• data-efficient learning
• image processing