← All projects

game · 2023

Maniacs

A gaze-driven horror game where looking can save you—or put you in danger.

My role
Core Unity Developer · Gameplay & Gaze Interaction
Project visualization of an abandoned moonlit town, with a distant figure and pools of warm light.

One light. Two very different risks.

Look to aim the flashlight independently of movement. Illumination freezes shadow enemies, but sustained light can awaken petrified ones. Limited batteries and gaze-activated puzzles make attention part of the survival loop.

Project visualization of a first-person flashlight illuminating a rune on a stone wall.
Project visualization of a hostile creature illuminated in a dark village street.

From webcam input to gameplay.

I connected GazePointer’s webcam input to independent flashlight aiming in Unity, then tied that input to enemy responses, gaze-duration puzzles and battery use. Profiling and playtest feedback guided iteration.

Project visualization of the GazePointer webcam-control panel and debug preview, with facial landmarks and gaze direction.
Project mechanism diagram: webcam gaze input maps to flashlight direction, which drives puzzles, opposing enemy responses and battery use.

Inside the prototype.

Project visualization of an enemy prefab in Unity with detection volumes, Animator, navigation and gameplay-script components.
Project visualization showing the Unity game view with inventory, battery, health and escape objectives.
Project visualization showing a top-down Unity editor view of the level, scene hierarchy and enemy prefabs.

More freedom. A precision trade-off.

Players completed a tutorial and play session with conventional or webcam-based gaze controls, followed by a five-point survey and interviews. The team’s survey summary covers 34 responses, 17 per condition.

Gaze input received higher exploration and challenge ratings; conventional controls were rated more accurate. These descriptive findings made calibration and clearer feedback priorities for the next iteration.

Playtest survey dashboard summarizing 34 responses across conventional and gaze-input conditions: gaze input rated higher for exploration, conventional input higher for flashlight accuracy.

Toward more accessible input.

Could a low-cost webcam lower the cost of gaze-assisted play for people who find hand controls difficult? This prototype explores that question; it does not establish that webcam tracking can replace dedicated eye trackers.

Next, I would test calibration, lighting tolerance and input recovery, compare accuracy and usability with dedicated hardware, and evaluate hybrid gaze/controller input with players who have limited hand mobility.

Skills & tools
  • Unity
  • C#
  • GazePointer
  • Input mapping
  • Gameplay systems
  • Unity Profiler
  • Playtesting
  • HCI
Context
Collaborative graduate project · Northeastern University · Boston, 2023