In this study, Camellia oil is co‐extracted from Camellia oleifera seeds and green tea scraps by supercritical carbon dioxide (SC‐CO2), which is optimized on the extraction yield, ABTS‐scavenging activity, and total polyphenols content (TPC) of oil by single‐factor experiments combined with response surface methodology (RSM). The extraction temperature, pressure, dynamic time, carbon dioxide (CO2) flow rate, and seed mass ratio were investigated with single‐factor experiments. The results indicated the optimum CO2 flow rate and dynamic extraction time were 15 L hour−1 and 60 min (i.e., 2.382 kg CO2/100 g sample). Furthermore, the complicated effects of extraction temperature (40–50 °C), pressure (20–30 MPa), and seed mass ratio (0.25–0.75) were optimized by RSM based on the Box–Behnken design (BBD). The models with high R‐squared values were obtained and used to predict the optimum operating conditions of the process. Under the optimum operating conditions (i.e., temperature of 46 °C, pressure of 30 MPa, and seed mass ratio of 0.35), the extraction yield, ABTS‐scavenging activity, and TPC of oil were 14.43 ± 0.17 g/100 g sample, 73.70 ± 0.34%, and 2.18 ± 0.05 mg GAE/g oil, which were in good agreement with the predicted values. In addition, the experiments indicated that the Camellia oil obtained was rich in polyphenols, resulting in better oxidation stability and antioxidant activity than the original oil.