Conventional dietary assessment methods rely heavily on self-reporting and are prone to errors. Thus, there is a growing need for more specific and accurate dietary assessment methods. Due to the technological proliferation, image-based smartphone apps with intelligent features, which may improve dietary assessment, have been developed. However, there is room for improvement in the field of mHealth due to the lack of validation and robust scientific work behind the use of such systems. Moreover, when using image-based nutrition apps, a large number of pictures (approx. 12%) is discarded due to human errors made in the capturing procedure. Trials should be conducted under free-living conditions and mHealth solutions should be compared with conventional ones. Collaboration of multidisciplinary teams is of vital importance and especially the needs of healthcare professionals and end-user should be taken into account when designing and developing nutrition apps.