Interdisciplinary research on autism diagnosis and training for the autistic children are current requirements. Behavioral analytics are the techniques followed by pediatricians for identifying ASD in children. However, the techniques successfully accepted by the pediatricians worldwide would be useful for the parents and pediatricians only for the elementary school-aged children, and the system is not helping for early diagnose among children of age 2. Machine learning, image processing, and pervasive techniques are trending for early diagnosis of ASD. Hence, the chapter has done a research-based survey and detailed analysis on behavioral analytics algorithms, neuro imaging, and eye tracking techniques for early diagnosis. The behavioral analytics survey mainly focuses on social and meta-cognitive skills, and the input dataset has been taken from parents, trainers, teachers, and peer groups. The chapter also identifies that computing technologies with neuro-imaging and pervasive techniques are better than the behavioral skill analysis for early diagnosis.