Why Is Reliable Autofocus Needed?
Reliable face‑tracking autofocus ensures that your subject stays sharp, even when they move within the frame.
If you plan on recording yourself while moving, you need a camera with responsive AF tracking. Since you cannot change focus points while filming yourself, eye or face AF becomes critical. Cameras with phase‑detect AF are especially effective here, as they are far more consistent for moving images compared to contrast‑detect AF, which tends to hunt and drift.
Pro‑level autofocus systems can track subjects across the frame by locking onto faces and eyes, making photography less arduous and significantly more creative.
Today, tracking, eye detection, and general autofocus performance are nearly flawless.
Capturing Moments: The Autofocus Process
Capturing a sharp moment follows a logical sequence:
Precompose: Select the focus point where you want the subject in your composition
Lock: Half‑press the shutter to start focusing on what’s under the selected focus point
Track: Follow the subject while maintaining composition, with AF assisting
Capture: Fully press the shutter when everything aligns
It looks easy—and it is—but only once you’ve mastered it.
This involves understanding:
Focus and recompose, or focus before composition
Setting other parameters like exposure in advance
Real‑World Focusing Challenges
Autofocus requirements change depending on real‑world situations:
A stationary subject: portraiture, e‑commerce, or landscape
A stationary subject within a group of similar subjects: group photos where focus is on a specific person
Moving subjects within a group: events such as weddings or functions
Subjects that appear and disappear unpredictably: weddings, functions, journalism
Fast‑moving subjects with unpredictable movement: sports and wildlife
Stand‑alone subjects with changing distance: vlogging or live event streaming
Each scenario demands a different autofocus strategy.
Why Does Commercial Cinema Still Use Manual Focus?
Commercial cinema continues to rely on manual focus for two main reasons:
High‑end productions can afford highly skilled professionals called focus pullers, whose role is critical because out‑of‑focus footage usually cannot be fixed in post‑production.
Cinematic environments are controlled—lighting, movement, framing, and retakes are possible. This margin for error does not exist in sports, wildlife, or journalism.
Manual focus is a skill that must be learned and practiced. While it’s easier with still objects, it becomes increasingly difficult with moving subjects.
Predictive Autofocus: The Core Idea
Manufacturers introduced predictive autofocus to reduce cognitive load on photographers.
The idea is simple:
Focus on a subject
The camera stays locked on it
If either the subject or photographer moves, the camera maintains focus
This allows creators to concentrate on composition and exposure instead of constantly managing focus.
Autofocus Modes Explained
Across brands (without naming modes differently), autofocus can broadly be divided into:
Stationary Subjects
AF‑S
Moving Subjects
AF‑C
Focus Area Types
Pinpoint
Single Point
Dynamic Area
Wide Area
In many modes, you first focus, lock focus, and then compose.
With AF‑S:
Focus locks when the shutter is half‑pressed
Focus priority ensures the camera won’t take a shot unless the subject is in focus
It can also be used to focus at a distance and wait for a subject to enter the frame
This makes techniques like focus‑and‑recompose reliable.
Speed vs Accuracy: The Debate That Ended
Autofocus technology once had to choose between being fast or accurate.
That debate is now over. All brands have chosen both.
However, you still need to:
Acquire initial focus quickly and accurately
Be comfortable with your gear
Use buttons instinctively without looking
In most real‑world situations, modern systems make it very hard to go wrong.
Autofocus Evolution: Problems & Milestones
DSLR Era
Fast AF achieved using a separate AF sensor below the mirror
In live view, the mirror flipped and contrast‑detect AF took over, which was slower
Cross‑type AF points were aggressively marketed
Subject identification existed but was limited
Few key innovations stood out:
Nikon 3D Tracking
Designed to remove precompose steps
Tracks subjects using color, contrast, and predictive algorithms
Popular for birds in flight and sports
Struggles when many similar subjects are present
Canon Dual Pixel Autofocus
The Mirrorless Era
Manufacturers attempted to keep phase‑detect AF active in video using translucent mirrors (e.g., Sony SLT series).
Eventually, removing the mirror allowed:
Massive numbers of phase‑detect AF points directly on the sensor
Elimination of optical viewfinders
The disappearance of the mechanical shutter sound
Example:
Sony A7 M5 with 759 PDAF points covering 94% of the sensor
AI and Subject Recognition
Parallel advances in AI introduced subject identification for:
Humans
Eyes, faces, heads
Animals and specific body parts
The goal was to eliminate precompose and manual focus point selection.
Sony Real‑Time Autofocus
Launched in 2019 with the Sony A6400, this replaced Lock‑On AF:
Prioritizes the eye
Falls back to face detection
Continues tracking even if the subject turns away or is briefly obscured
Canon Eye Control
Tracks the photographer’s iris to move focus points
Nikon 3D Tracking (Modern)
Improved usability through higher AF point density
Automation vs Control
None of these systems have become default modes because automation is not always desired.
While automation simplifies the process, high‑control photographers may see this as a loss of absolute control.
Modern mirrorless cameras can track eyes faster than humans can move—but mastery still requires understanding:
How much control to give the camera
When to override automation
How practice builds confidence
Increasing automation offers benefits, but also introduces complexity and confusion.
Final Thought
To fully master a camera, you must understand and practice its focusing system. Many photographers may not have the time or inclination to master everything—but choosing the right level of automation requires understanding how autofocus truly works.