Channel selection, quantization, and compression are used to reduce data size of textures used in pixel correction. For example, an apparatus such as a head-mounted display may include circuitry that retrieves a compressed texture from the memory, the compressed texture being generated using various compression techniques, and decompresses the compressed texture to determine adjustment quantization values for sub-pixels based on the compressed values. The circuitry determines reconstructed brightness adjustment levels for the sub-pixels based on the adjustment quantization values, and renders an image frame based on the reconstructed brightness adjustment levels. In some embodiments, the apparatus or a separate device generates the compressed texture in a calibration and stores the texture in a memory of the apparatus for use during the pixel correction.