As a kind of global epidemic, obesity is closely related to many Non-Communicable Diseases (NCDs), such as cardiovascular disease, chronic obstructive pulmonary disease, type 2 diabetes and even many specific types of cancer. In addition, obesity is also associated with poor cognitive function and neuro-degenerative disorders, such as dementia. The proportion of adults who were overweight or obese worldwide have increased dramatically over the past decennia. Compared with countries such as the UK and USA, China had a lower rate of obesity in adults, but the absolute number of obese people in China is exceeded only by that in the USA. Neuroimaging of neural responses in those with obesity or who are overweight has largely been conducted using food images as opposed to actual food ingestion, and a meta-analysis of these studies would provide a better understanding of the neural mechanisms underlying the development of obesity. Therefore, the present study systemic reviewed existing literature on functional and structural imaging researches of obesity. Evidences from the functional imaging researches, which showed that the existence of neural differences in response to food cue processing between healthy weight and obese subjects. Differences were identified in two brain circuits: (1) limbic and paralimbic areas associated with salience and reward processes, such as amygdala, insula, orbitofrontal cortex and parahippocampal gyrus; (2) prefrontal areas supporting cognitive control processes, such as lateral prefrontal and anterior cingulate cortex. The hyper-responsivity of reward-related areas, the hypo-responsivity of cognitive control-related areas, together constitute the susceptible factors of obesity. There is also evidence that obesity is related to structural brain differences. Evidences from the structural imaging researches, which showed that obesity is associated with brain atrophy. For example, higher BMI are linked with low global grey matter volume as well as with regional grey matter (GM), include left orbitofrontal, right inferior frontal, parahippocampal, fusiform, lingual gyri, right cerebellar regions and right precentral gyri. In addition, increasing BMI was independently associated with lower fractional anisotropy (FA) in the genu, splenium, and fornix. All this evidence provided anatomical evidences from different perspectives of obesity. Finally, we suggest that the following aspects might shed light on future research: (1) more long-term longitudinal studies, only in this way can we confirm whether the neural differences in response to food and structural brain differences predicts obesity in the future, particularly after controlling for covariates such as socioeconomic status, age, sex, inflammation, nutritional habits, among others; (2) the expansion of the intervention strategy, for example, cognitive remediation therapy, designed to improve neurocognitive abilities such as attention, memory and executive function, which has been shown to be useful to treat cognitive deficits in anorexia nervosa. Hence, by increasing the levels of cognitive function this may aid in helping individuals make suitable lifestyle decisions, stop the vicious cycle and maintain it long term; (3) integration of multi-modal brain imaging data, such as, diet and lifestyle are the most readily identifiable causes of obesity, yet it is highly heritable, so it is necessary to determine if the neuroanatomical pathways compromised by this obesity-risk gene themselves indicate a mechanistic pathway for obesity.