Emerging Technologies and Applications for Computer Vision and Recognition Systems

Dear Colleagues,

Computer vision (CV), the science of teaching machines to understand our visual world, has witnessed a paradigm shift in the past decade from hands-on methods to the use of deep neural networks. Deep learning has significantly boosted performance in a broad range of CV tasks, such as classification, detection, and segmentation.

In essence, deep learning can generalize superior networks and lead to an improved performance, which can be attributed to the use of a large number of labeled data, such as ImageNet. However, there may not be enough labeled data available for some applications (e.g., medicine, military), and thus the robustness and generalization of networks cannot be guaranteed. Inspired by the impressive ability of humans to learn new tasks without large amounts of prior knowledge, zero and few-shot learning have been introduced to address this challenge. Additionally, novel technologies and applications have emerged with significant technological breakthroughs in CV and Natural Language Processing (NLP), such as Transformer and Contrastive Language-Image Pre-Training (CLIP). For example, multi-modal learning (e.g., image/text, image/video, text/audio) has been employed to transfer semantic information from one modality to another, which is an essential sub-area of CV. Additionally, transformer-based pre-trained models have significantly boosted the performance of various CV applications, such as image retrieval, semantic segmentation, image generation, and video generation.

Despite the surge of research interest in this area, several core issues in CV, such as cross-modal analysis, zero/few-shot learning in detection or segmentation, and other applications, remain open problems. This Special Issue aims to provide a platform for original contributions discussing theories, algorithms, model architecture design, and novel CV applications. Topics of interest in this Special Issue include, but are not limited to:

• Multi-modal analysis (e.g., retrieval, grounding, reasoning, generation, understanding, pre-training);
• Zero/few-shot learning (e.g., un/self/semi/weak supervised, cross-domain);
• Object detection and tracking (e.g., 2D/3D, remote-sensing-related);
• Image segmentation (e.g., pixel-based/label-based/semantic segmentation);
• Applications (e.g., agriculture, weather, healthcare).

Prof. Dr. Zhong Ji
Dr. Yunlong Yu
Prof. Dr. Lei Chen
Guest Editors

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 250 words) can be sent to the Editorial Office for assessment.

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.