Person Re-identification is defined as a recognizing process where the person\nis observed by non-overlapping cameras at different places. In the last decade,\nthe rise in the applications and importance of Person Re-identification for\nsurveillance systems popularized this subject in different areas of computer\nvision. Person Re-identification is faced with challenges such as low\nresolution, varying poses, illumination, background clutter, and occlusion,\nwhich could affect the result of recognizing process. The present paper aims to\nimprove Person Re-identification using transfer learning and application of\nverification loss function within the framework of Siamese network. The Siamese\nnetwork receives image pairs as inputs and extract their features via a\npre-trained model. EfficientNet was employed to obtain discriminative features\nand reduce the demands for data. The advantages of verification loss were used\nin the network learning. Experiments showed that the proposed model performs\nbetter than state-of-the-art methods on the CUHK01 dataset. For example, rank5\naccuracies are 95.2% (+5.7) for the CUHK01 datasets. It also achieved an\nacceptable percentage in Rank 1. Because of the small size of the pre-trained\nmodel parameters, learning speeds up and there will be a need for less hardware\nand data.\n