Efficient and high quality texture compression is important because computational powers are always limited, especially for embedded systems such as mobile phones. To access random texels instantly, down sampling and S3TC are the most commonly used methods due to the fixed compression ratio at every local region. However, the methods are content oblivious, which uniformly discard texel information. Therefore, we present content aware warp to reduce a texture resolution, where homogeneous regions are squeezed more to retain more structural information. We then relocate texture coordinates of the 3D model as the way of relocating texels, such that rendering the model with our compressed texture requires no additional decompression cost. Our texture compression technique can cooperate with existing methods such as S3TC since the compression strategies are independent. By reducing texture resolution in a content aware manner, followed by texture compression using S3TC, as the experiments show, we achieve higher compression ratios without degrading visual quality significantly.