This lecture offers an end-to-end tour of image and video coding from why it matters to how it works and where it’s going. It opens with real-world applications and the constraints they impose (bitrate, latency, power, quality, HDR, depth). It then establishes core notions, including sampling and color spaces, spatio-temporal redundancy, perceptual versus signal fidelity, rate–distortion theory, and objective versus subjective quality assessment.
The technical backbone encompasses the coding toolbox, which includes transforms (DCT, wavelets), intra/inter prediction and motion compensation, quantization, entropy coding, in-loop filtering, rate control, scalability, and perceptual tools. A comparative study of standards follows with emphasis on the JPEG family (baseline JPEG, JPEG 2000, JPEG XR, JPEG XL, and a look at JPEG Pleno/AI trends) and the MPEG video lineage (MPEG-1, MPEG-2, AVC/H.264, HEVC/H.265, VVC/H.266), highlighting design choices, performance trade-offs, and interoperability.
In addition to lectures, students complete mini-projects that implement or experiment with specific components or codecs.
- Professor: Touradj Ebrahimi
