Is AI Face Replacement Transforming TV Production?
Blurred faces on television have always been frustrating. They’re practical, necessary, and dull, of course. But in protecting people’s privacy, you miss the important parts of any impactful and emotionally-charged story — people’s reactions.
Recently, though, something really interesting happened as we were watching Channel 4’s latest 24 Hours in Police Custody. Instead of using the usual blurred faces to protect children’s identities during particularly sensitive scenes, they opted to use AI face replacement instead.
The result? We could suddenly see nuanced facial expressions; fear, uncertainty, relief. All without compromising identity. This relative new approach left we wondering how AI is quietly reshaping the boundaries of storytelling methods in film and television.
What is AI face replacement?
AI face replacement is the use of using “artificial intelligence” to swap one person’s face with another in a photo or video, while keeping their expressions, lighting and facial expressions authentic.
At its core, it works like this:
- Analyse the face: AI learns the structure of the face such as distance between eyes, face structure, expression used etc.
- Map movements: It then begins to track how the face moves over time, frame by frame.
- Generate the replacement: Using the motion data, the AI now has a framework in which to render another face over the top so it matches the face perfectly.
- Blend seamlessly: The final output is composited so it looks natural and part of the scene.
It might sound like it’s not a new thing, and that’s because you’re right. Over the years there have been lots and lots of variations of digital face replacements technology, such as Robert DeNiro’s de-aging in The Irishman, or Oliver Reed’s replacement in Gladiator.
However, with the development of AI technology, digital face replacement is likely to become far more commonplace as what previously needed huge spending budget to achieve a decent result, now has a huge reduction in cost with AI tools now at our disposal.
Why TV producers are embracing AI
The technology surrounding AI’s appeal is clear — it’s creating the balance of emotional depth and identity protection that makes AI face replacement an attractive choice for production companies that want to enhance their film and television shows. Producers want their viewers deeply connected to stories without breaching any privacy or ethical boundaries, and using AI is great way to achieve this at a low cost.
And it makes sense, right? If you’re watching a TV show, you can naturally interpret emotional cues from facial expressions, which allows you to form a deeper emotional connection with the subjects on screen, rather than watching a gigantic blur prevent every subtle emotion or reaction being visible and understood by us. Not only that, but the clarity of using AI face replacement enhances accessibility, especially for viewers relying on lip-reading or other visual cues.
Detailed storytelling becomes possible, capturing critical emotional reactions even during highly sensitive situations without compromising privacy. Additionally, clear archival footage offers valuable historical documentation, preserving authentic emotional context for future reference without the ambiguity caused by blurred identities.
It goes beyond replacing blurs
As slightly touched upon earlier in the blog, AI face replacement isn’t just an opportunity to remove blurs and replace them with realistic faces. It goes far beyond that, stemming into many other use cases.
An example of such would be enhancing old historical footage that used low resolution image capture, or perhaps a production that used film stock that degraded over time. You might be able to slightly make out faces, or even general moving images as the show goes on, but as our tv viewing experience has now evolved into 4K and 8K image resolution, it’s far from optimal! So, applying some AI technology to interpret image data and replace it with a higher-resolution replacement is an interesting application.
The BBC recently highlighted this in a case study exploring AI-generated faces within war-crime documentaries, showcasing the delicate balance between emotional authenticity and ethical storytelling. This groundbreaking approach illustrates AI’s potential to elevate impactful storytelling while safeguarding vulnerable individuals.
Looking ahead
AI face replacement is quickly evolving from an intriguing novelty into an industry standard. It’s changing how we tell stories, challenging ethical frameworks, and pushing regulatory boundaries.
At DAKONA, we’re continually exploring these innovative techniques. It’s more than technology — it’s about telling powerful stories ethically and effectively. So, next time you notice subtle, emotional reactions clearly visible on a documentary, remember: there might be more AI involved than you realise.
Interested in AI for video production? Drop us a line! We’d love to chat.