The aim of this study was to assess the accuracy of self-estimated vegetable and whole grain serving sizes in a self-served buffet meal. The study took place in a laboratory setting where an Intelligent Buffet was used to register the exact weight of each food type that was self-served by each participant. The initial sample consisted of 58 participants recruited from Aalborg University in Copenhagen, of which 52 participants (59% male) provided complete estimates on the weight of whole grains and 49 participants (63% male) provided complete estimates on the weight of vegetable servings in their meal. The majority of the participants were students aged 20-29 years (85% for whole grain responses and 82% for vegetable responses). Significant differences between self-estimated and actual portion size estimates were observed for both vegetables and whole grains (P < 0.001). The mean self-estimated weight of a vegetable serving was 218(±134) g compared to the mean actual weight of 74(±44) g. The mean self-estimated and mean actual weights of a whole grain serving were 36(±34) g and 10(±9) g, respectively. There was no significant correlation between self-estimated and actual weights for each food group (P > 0.05). In conclusion, the participants' ability to accurately assess the serving size of vegetables and whole grains in a self-served meal did not correspond with the actual amount served. This may have implications for consumer interpretation of dietary recommendations used in nutrition interventions in Denmark.