Indoor device-free localization (DFL) using radio frequency (RF) signals has many important applications including navigation, security and inventory control. One of the most common approaches to indoor DFL is Radio Tomographic Imaging (RTI) which is based on line-of-sight (LOS) signal models and often utilizes WiFi signals. In this work, we propose a new extended Rytov phaseless imaging (xRPI) technique for indoor imaging that can image the shape and refractive index of objects in the indoor environment using WiFi RSSI measurements. Estimation of refractive index can be used to distinguish different object types within the environment. In addition, methods to reduce the number of RSSI measurements required for indoor imaging are also developed and these use a strong spatial sparsity constraint which directly arises from imaging change in the environment. Theoretical developments that support these contributions include increasing the validity range of the Rytov approximation by using a high frequency approximation to the total field, incorporating temporal background subtraction, incorporating electric field sensing using directional antennas, and utilization of Fresnel zones. Experimental demonstration of the performance of the proposed method is provided using a 20 node WiFi system. The results show that xRPI provides considerable improvement over RTI in accurately imaging shapes of different objects in an indoor environment. It also provides an estimate of the refractive index which is not possible with RTI or other Wi-Fi based indoor imaging techniques.