Ultrafast optical imaging plays a key role in capturing transient time-evolving phenomena occurring on extremely fast timescales, driving breakthroughs in fields ranging from physics to biology. Conventional ultrafast imaging techniques, such as compressed ultrafast photography (CUP), can record transient events at hundreds of trillions of frames per second using chirped ultrashort pulses. However, these systems are usually large in size, limiting their versatility and practical use. How can ultrafast imaging systems be made more compact, reliable, and versatile without sacrificing performance? To address this issue, this study proposes integrating advanced nanophotonics, compressed sensing, and deep learning to develop single-exposure compressed ultra-compact femtosecond photography (CUF). The simulation shows that by utilizing a super-dispersive metalens, CUF combines focusing and dispersion functions in a compact device to achieve single-exposure ultrafast imaging, providing a transformative solution to overcome the limitations of conventional CUP technology in terms of system size. This work was recently published in Ultrafast Science and featured as the cover story.