To efficiently utilize the complementary attributes in RGBT images, we proposes an object tracking algorithm called Isomeric Feature Encoding Network (IFENet). Based on the different characteristics of RGBT images, IFENet employs the global-memory enhancement (GME) in the early stage of image feature encoding to explore detailed information (such as texture and color) in the RGB modality. It also utilizes the border-region salience enhancement (BRE) to improve the saliency difference between the object region and the background. Furthermore, an interest region sampling is introduced to reduce computational consumption and improve the operational efficiency. Validation results on the open-source datasets demonstrate the effectiveness of IFENet. Compared to current mainstream RGBT tracking algorithms, IFENet achieves better tracking accuracy and robustness. It can effectively handle challenging scenarios such as fast-moving objects, large-scale deformations, and camera motion. Moreover, IFENet achieves an average tracking speed of 62FPS, meeting real-time tracking requirements.