Several seminal papers have significantly shaped the field of computer vision. One of the most influential is "A Computational Approach to Edge Detection" by John Canny (1986), which introduced the Canny edge detector, a crucial method for detecting edges in images. This paper laid the groundwork for many subsequent edge-detection algorithms. Another important paper is "Object Recognition from Local Scale-Invariant Features" by David Lowe (1999), which introduced the SIFT (Scale-Invariant Feature Transform) algorithm. SIFT is widely used for feature extraction in object recognition, particularly in tasks where scale and rotation variance are significant. A foundational paper in the deep learning era is "ImageNet Large-Scale Visual Recognition Challenge" by Olga Russakovsky et al. (2015), which detailed the ImageNet dataset and the deep learning methods used for image classification. This paper is credited with demonstrating the effectiveness of Convolutional Neural Networks (CNNs) in large-scale image classification tasks. Another key paper is "Fast R-CNN" by Ross B. Girshick (2015), which improved object detection by integrating region proposal networks with CNNs. These works, among others, continue to influence modern computer vision techniques.
What are the seminal papers on computer vision?

- Master Video AI
- The Definitive Guide to Building RAG Apps with LangChain
- Embedding 101
- Large Language Models (LLMs) 101
- Vector Database 101: Everything You Need to Know
- All learn series →
Recommended AI Learn Series
VectorDB for GenAI Apps
Zilliz Cloud is a managed vector database perfect for building GenAI applications.
Try Zilliz Cloud for FreeKeep Reading
How do you automate data analytics workflows?
Automating data analytics workflows involves using tools and techniques to streamline the process of collecting, process
What are target networks in DQN?
Target networks in Deep Q-Networks (DQN) are a crucial element used to stabilize the training process of the Q-learning
How does quantum computing help solve optimization problems faster than classical systems?
Quantum computing offers a different approach to solving optimization problems compared to classical systems, primarily