Solar cells are mainly used as power sources, but can be used for sensing as well. We propose a novel indoor system that exploits solar cells to track people by monitoring the changes in light intensity caused by their shadows and reflections as they walk by. Our framework has three main components. First, we develop a simulator based on a ray-tracing model to determine how the solar cells should be positioned in the tracking environment to maximize the signal to noise ratio. Next, we apply changepoint detection methods to convert the (noisy) solar cell signal into a binary detection signal. Our detection method uses a Bayesian approach, which allows our system to work well in various environments, with natural and artifical light. Finally, the binary output from multiple solar cells is fused to track multiple targets. The tracking engine is based on a particle filter implementation based on the probability hypothesis density filter. This approach allows us to perform tracking without knowing the actual number of targets in the environment. To evaluate our framework, we build small tags that consist of a solar cell, a micro-controller and a wireless module, and deploy them in a real apartment. Ours results show that our system allows solar cells to track people under different lighting conditions, during day and night.