Objective Verify the effect of non-periodized and linear periodized combined training on body image perception and body dissatisfaction in adults with obesity. Design A randomized clinical trial on the effect of two types of combined training periodization was carried out on people with grade I and II obesity. Setting Participants were allocated into three groups: non-periodized periodization group, linear periodization group, and control group, with 23 participants in each. Subjects Adults with obesity, with a body mass index between 30 kg/m² and 40 kg/m². Interventions The intervention lasted 16 weeks (separated into three mesocycles of 4 weeks and 1 week of familiarization), in 3 weekly sessions of 1 hour each, composed of aerobic (30 min) and muscle strength exercises (six exercises) in the same session. The group with non-periodized model maintained uniform intensity, volume, and workload throughout the mesocycles. The group with the linear periodization model started with low initial intensity, subsequently introducing a gradual increase of intensity in the mesocycles. Main measures Body image perception (current and ideal silhouette) and body dissatisfaction were assessed using the Stunkard silhouette scale. Results The perception of the current silhouette increased only in the control group (Δ = 1.16; p = 0.04). There was a significant reduction in the ideal silhouette over time ( p = 0.001), especially in non-periodized group (Δ = −1.26). Body dissatisfaction increased significantly only over time ( p = 0.001), especially for control group (Δ = 1.47). Conclusions Regardless of periodization, combined training effectively maintained the perception of the current silhouette and reduced the ideal silhouette in adults with obesity. Trial registration Brazilian Registry of Clinical Trials (RBR-3c7rt3).