In one embodiment, a method includes rendering, for a VR display device and based on a field of view of a user, a first output image of a VR environment comprising a virtual boundary corresponding to a real-world environment; determining, responsive to the user approaching within a first threshold distance of the virtual boundary, a speed of movement and the field of view of the user; accessing one or more images of the real-world environment captured by one or more cameras of the VR display device; and rendering, for the VR display device, a second output image comprising a portion of the VR environment and a portion of a passthrough view of the real-world environment based on the accessed images, where an area of the portion of the passthrough view is based on the determined speed of movement and the field of view of the user.