Why did JXL perform worse than WebP on this corpus?

There are numerous factors that affect how well these image formats compress data can be used to explain why JPEG XL (JXL) performs worse than WebP on a given batch of photos. Based on the characteristics of the image corpus and the chosen compression settings, the efficacy of image compression formats can vary.

There are various reasons why JXL might have produced less desirable outcomes than WebP in this particular case:

1. Configuration for Compression

The precise compression settings used have a significant impact on the effectiveness of JXL and WebP compression. JXL settings that were not tailored to the distinct qualities of the photographs in the corpus could produce less-than-ideal compression results.

For WebP, the same rule is true.

Different image formats were created with different types of material in mind. For example, WebP is recognized for how well it handles photographic images with vibrant color palettes and gradients. WebP may have had an advantage if such photos made up the majority of the corpus.

2. Complexity of Images

A wide range of picture types, including those with complicated patterns and high quality, can be supported by JXL. However, because of its specialized compression techniques, WebP might have provided better performance if the corpus had included photos with intricate or highly detailed patterns.

3. Image Colour Depth

Compression results can be considerably impacted by an image’s color depth. The performance disparity could be explained if the corpus included images with a specific color depth that is more compatible with WebP’s compression methods.

4. Lossless vs. Lossy Compression

JXL has a lossless compression mode, but WebP generally uses lossy compression. Lossless vs. Lossy Compression. JXL may have performed worse than WebP if the corpus required lossless compression to preserve image quality but was used in lossy mode instead.

5. Optimization and Fine-Tuning

The degree of optimization and fine-tuning for particular activities can have an impact on the image compression formats. WebP might perform better than JXL if it has undergone considerable optimization specific to the image types in the corpus.

6. Implementation of the encoder

Depending on the specific implementation, image encoder performance may vary. Results may vary depending on encoding schemes, optimization methods, and program versions.

7. Subjective Evaluation

The way users judge the quality of the images can sometimes affect how objectively an image compression test is conducted. What one person would find to be acceptable quality, another might find to be poor. Therefore, human preferences can affect how well picture formats are perceived to perform.

In conclusion, the properties of the image collections the compression settings, and the particular requirements of the task at hand are just a few of the variables that affect how well JXL and WebP perform in comparison to one another.
It is important to analyze these parameters and possibly experiment with other compression settings or formats to reach the required results for the particular image collection under consideration in order to understand why JXL produced less favorable results than WebP in a given circumstance.

May these links help you out more: