The Camera-on-a-Chip Revolution: How Dr. Eric Fossum’s CMOS Invention Transformed the World, One Photon at a Time
In an era where high-resolution cameras are so ubiquitous that they sit in our pockets, doorbells, and dashboards, it is easy to forget that the technology enabling this visual revolution is barely a few decades old. At the heart of this transformation is the Complementary Metal-Oxide-Semiconductor (CMOS) image sensor, an invention dubbed the “camera-on-a-chip” that has fundamentally changed how humanity captures, shares, and understands the visual world. The story of this breakthrough traces back not to consumer electronics, but to NASA’s ambitious quest to install small, low-powered, high-resolution cameras on interplanetary spacecraft. Dr. Eric R. Fossum, the inventor of this technology, is the man who fathered this era of space-based photography and, inadvertently, the vast ecosystem of micro-cameras that now pervade modern life. Before the CMOS sensor, photography relied on either bulky photographic film or the more advanced Charge-Coupled Devices (CCDs). The Apollo 11 mission famously utilized purpose-built Hasselblad cameras, which were so heavy that astronauts left them on the lunar surface, returning to Earth with only the film reels for development. CCDs, the immediate precursor to Fossum’s work, represented a leap forward—they detected photons via individual pixels that fed an electrical charge to a single readout amplifier in a row-by-row configuration. While these sensors reduced weight and produced more vivid imagery than film, they were notoriously power-hungry and generated significant heat, limiting their practicality for portable and battery-operated devices.
The CMOS image sensor improved dramatically on the CCD design by giving each individual pixel its own local conversion circuit and amplifier, allowing every pixel to be read simultaneously. This architectural leap drastically reduced the energy required per image and minimized heat generation, making high-quality digital imaging viable for mass-market applications. Today, CMOS sensors form the backbone of nearly all modern cameras, from the lens in your smartphone to the Advanced Driver Assistance Systems enabling self-driving cars, and even the sophisticated imaging instruments still being sent into deep space. The technology’s journey from a NASA laboratory to the center of global communication is a testament to the power of government-funded fundamental research. Fossum, who developed the sensor while working at NASA’s Jet Propulsion Laboratory (JPL), recently sat down for an interview to reflect on the invention’s enormous societal footprint, his work on the next generation of imaging technology, and the ethical dilemmas that come with creating a tool that sees everything. When asked if he ever anticipated the profound impact his invention would have, Fossum admitted to a mix of prescience and humility: he thought it could and should become ubiquitous, but his confidence that it would actually happen was low due to “all kinds of unknowns.” Crucially, he did not foresee the social-media revolution, nor the rise of citizen journalism, nor the technology’s role in social justice movements—both in protecting good actors in law enforcement and identifying bad actors. He also acknowledged a darker side: “I also knew there could be impact on privacy and liberty, and I am uncomfortable with how that has evolved.”
The connection between NASA and digital imaging is deeper than many realize. JPL, Fossum noted, has been associated with cutting-edge digital imaging since its earliest activities in robotic space exploration. Engineers at JPL found innovative ways to beam images back from distant planets using digital data transmission, and as early as 1961, Eugene F. Lally at JPL proposed the idea of a “mosaic” image capture, referring to an array of pixels—the conceptual precursor to modern image sensors. JPL developed numerous digital imaging technologies designed to operate with high precision and reliability in the harsh environment of space, spanning a larger range of wavelengths than consumer products. Fossum’s job at JPL was specifically created to grow the lab’s bench-depth in CCD technology, which eventually led to his pivotal CMOS invention. This lineage came full circle during the recent Artemis II mission, which saw humans return to lunar orbit. Fossum followed the mission’s visual journey avidly, from iconic photos to crew-produced videos. While his initial thoughts were focused on the impressive imagery and his pleasure at humanity’s return to the Moon, he was struck by a beautiful irony: the NASA-invented image sensor technology, now employed in consumer smartphones and digital cameras, was itself used on the NASA mission, “further completing its promise to make cameras faster, cheaper and better.” The invention has effectively closed a loop, with space exploration creating a technology that now enables that same exploration to be documented by the public.
While the CMOS sensor has already changed the world, Fossum is not resting on his laurels. His current research at Dartmouth College centers on the Quanta Image Sensor (QIS), a radical new approach that he describes as creating “great images one photon at a time.” The QIS concept is technology-agnostic, but its implementation represents a fundamental leap in sensitivity. In the CMOS QIS, ultra-low readout noise makes it possible for the first time to discern individual photons via their generated photoelectrons, and to count with high accuracy the number of photoelectrons at each pixel. This remarkable feat is achieved at room temperature, at low voltages, and without the use of avalanche gain. The CMOS QIS developed at Dartmouth and later at the spin-off company Gigajot uses intra-pixel charge transfer—a technique invented at JPL—and extremely low readout-node capacitance to generate a large enough voltage signal from a single photoelectron to overcome background thermal noise in silicon. This allows for photon-counting capabilities that were previously only possible in bulky, cryogenically-cooled laboratory equipment. Alternative implementations are also emerging: Single-Photon Avalanche Photodiode (SPAD) technology has advanced significantly over the past two decades and is becoming useful for image sensor implementation. A SPAD QIS offers larger pixels and consumes much more power than the CMOS QIS, but provides time-tagging of photon arrival that is far more accurate, and it is being produced as a foundry process easily integrated with CMOS circuits. This next generation of sensors promises to push the boundaries of low-light imaging, potentially revolutionizing fields from astronomy to biomedical imaging.
Fossum’s professional journey has been deeply intertwined with the IEEE, an organization he joined as a student and has since graduated to Life Fellow. He emphasized that professional societies are vital for disseminating leading research among experts, noting that he has published many papers under the IEEE in the sub-silos of Electron Devices and Solid-State Circuits. He has benefited immensely from IEEE meetings such as the International Electron Devices Meeting (IEDM) and the International Solid-State Circuits Conference (ISSCC), and the IEEE was a partner in the early days of what is now known as the International Image Sensor Workshop, which he and others founded in 1986. Looking to the future of consumer image sensors, Fossum sees continued innovation in the near term through 3D integration and stacked wafer devices. The on-chip integration of more advanced functions—possibly including smart-image-sensor functions or edge processing—offers new avenues for improving image capture and understanding right at the focal plane. This is a dream he has harbored since the 1980s, when he was exploring smart image sensors and focal-plane image processing, though he admits the technology was too early in those days. When asked about Processing-in-Sensor (PIS), which fuses compute, the image sensor, and memory into one holistic edge system, Fossum noted that he has been thinking about this topic since his graduate school days. However, he cautioned that the acronym might be a poor choice and that the approach is highly application-dependent. He identified a key data communication and parallel processing problem: local pixel computing is easy, but global image computing—where different regions of an image have to be analyzed together—remains difficult on the focal plane. This challenge may lead to partitioning imaging systems into two parts, keeping local computing on the focal plane while offloading regional or global computing to avoid adding extra power dissipation to the sensor chip.
Despite his monumental achievement, Fossum remains remarkably grounded and collaborative. He was recently awarded the prestigious IEEE Jun-Ichi Nishizawa Medal, and he reflected on the honor with characteristic humility. “Gosh, who doesn’t like being recognized by your peers for your accomplishments?” he said. “Still, it is a very humbling event and makes me reflect on the great people that I have had the chance to work with over the years.” He was quick to credit the broader community, noting that the amazing CMOS image sensor technology we all use daily was made possible by the hard and innovative work of thousands of image sensor engineers around the globe. This collaborative spirit extends to his current daily routine. When asked about a critical tool in his workflow, Fossum surprisingly highlighted AI tools, which he considers great time-savers in finding and curating information. He frequently uses Microsoft’s Co-Pilot, learning from its apparent knowledge and sometimes from its insights. Yet, true to the thoughtful nature that has guided his career, he voiced a cautionary note: “I am always concerned with whom, directly or indirectly, I am sharing information and ideas with. I think this is something we should all ponder and ask hard questions of the AI giants.” As the world he helped create grapples with the implications of ubiquitous surveillance, AI, and data privacy, Fossum’s journey from JPL to the pinnacle of imaging technology serves as a powerful reminder that the most transformative inventions often arise from humanity’s most ambitious pursuits—and that with the power to see everything comes the profound responsibility to consider what we are looking at, and who is looking back.