In this episode of Computer Vision Decoded, Jonathan Stephens and Jared Heinly discuss different types of cameras used in 3D reconstruction, starting with smartphones and moving through mirrorless cameras, action cameras, drones, 360 cameras, fisheye cameras, event cameras, and thermal and multispectral cameras. They discuss the strengths and weaknesses of each type, including sensor size, lens distortion, programmability, and dynamic range, and provide advice for beginners on choosing a camera and learning 3D reconstruction, emphasizing starting with the camera you already own, typically a smartphone, due to its ease of use and accessibility. They also touch on the importance of understanding camera settings, lens choices, and capture techniques for achieving optimal results in photogrammetry and Gaussian splatting.
Sign in to continue reading, translating and more.
Continue