Third-person

Third-person
Load video YouTube YouTube might collect personal data. Privacy Policy ContinueDismiss
Information
Type	Seated, Standing, Room-scale
Required Devices	Gamepad
Description	player's POV trails comfortably behind their character, works for third-person games with gamepads
Pros	very easy to implement, no need for motion-tracked controllers
Cons	Less immersive, can cause VR sickness
Examples	Lucky's Tale, Chronos, Moss, Astro Bot Rescue Mission

Third-person is a Locomotion Method.

In a third-person virtual reality (VR) game or application, the user does not see the world through the eyes of the character they control. The virtual camera sits outside that character, usually behind or above it; the user steers the character with a Gamepad or other controller, while head tracking still decides where the user looks. Kent Bye of Road to VR wrote that placing the camera in the third-person perspective with finely tuned movement algorithms "solves a lot of the problems of VR locomotion." Third-person VR titles include Lucky's Tale and Chronos, both released with the Oculus Rift in March 2016, and the PlayStation VR games Moss and Astro Bot Rescue Mission.

Research comparing first-person perspective (1PP) and third-person perspective (3PP) in head-mounted display VR has found tradeoffs. Third-person views have been linked to better awareness of the surrounding space and, in one game study, to less simulator sickness; first-person views produce a stronger sense of owning and being located in the virtual body.

Reviewed 27 September 2026. Checked every game fact, developer quote, release date and research-paper finding in the new text against the cited reviews, developer blog, podcast, GDC Vault session and paper abstracts or full texts. About review dates.

Additional Information

In Third-person VR platformers such as Lucky's Tale, player's POV trails comfortably behind their character. Most of third-person games play well with gamepads as input methods.

Lucky's Tale, developed by Playful, was a bundled launch title for the Oculus Rift when the headset launched on 28 March 2016. Bye wrote that it "proved to a lot of VR developers that not only could a third-person perspective work, but that it could work so well as to be able to cause a fit of VR giggles." The game places its action within what he called "the near-field sweet spot of VR," about an arm's length from the player, and it was rated "Moderate" for comfort.

How it works

Third-person VR designs differ mainly in how the camera follows the controlled character:

Camera approach	How it behaves	Example
Trailing follow camera	The viewpoint follows the character through the level as the player steers it, with the action kept at close range.	Lucky's Tale
Forward-travelling camera	The player moves forward through each level on rails, following behind the character, rather than scrolling sideways.	Astro Bot Rescue Mission
Fixed camera per scene	The viewpoint stays still while the character crosses a diorama-like scene; the view fades to black and a new scene loads between sections.	Moss
Static cameras with transition volumes	A set of stationary cameras that the player looks around with head movement; walking through a trigger volume instantly switches to another camera.	Chronos
Stationary single-room camera	One stationary camera overlooks a single room; the screen fades to black between rooms.	Herobound: First Steps (Gear VR)

The player can also be present in the world as a character. In Astro Bot Rescue Mission the camera-tracked PlayStation 4 gamepad is always shown inside the game and doubles as a gadget such as a grappling hook, and the player headbutts certain obstacles. Road to VR described the game as embodying the player "as not just a camera but an actual character that's present inside the virtual world." Creative director Nicolas Doucet told GamesBeat that the team "settled for a game based 80 percent on controlling a character, and punctuated with first-person interactions here and there where it made most sense." Moss casts the player as a spirit that characters call "The Reader." The player guides the mouse Quill with the PS4 controller, reaches into the scene with the motion-tracked controller to move parts of the world, and looks around with the headset's tracking to find better angles on hidden paths.

Use in games

Title	Developer	Platform	Release	Camera
Lucky's Tale	Playful	Oculus Rift	28 March 2016	Follow camera with action kept at about arm's length; bundled with the Rift at launch
Chronos	Gunfire Games	Oculus Rift	28 March 2016	Camera locked at a set point in each room; static head-controlled cameras switched by transition volumes
Moss	Polyarc	PlayStation VR	27 February 2018	Static third-person view of diorama scenes
Astro Bot Rescue Mission	Japan Studio (Team Asobi)	PlayStation VR	2 October 2018	Player travels forward on rails behind the character; 26 levels

Gunfire Games' earlier Gear VR titles Herobound: First Steps and Herobound: Spirit Champion used a stationary camera overlooking a single room. Design director John Pearl compared the result to "looking into a shoebox diorama" and said the team found it "made for a very comfortable VR experience." For Chronos the studio wanted combat that could spill from one room into the next and spaces where players could look back at where they had been, which led to its multi-camera system. Pearl wrote that some may think a third-person point of view takes away from VR, but that the player "still feels a great sense of presence" in the world.

Polyarc made player comfort and approachability its highest priority for Moss. In an XRDC 2018 session, the studio said this led it to build the world from "diorama-size puzzles you interact with in 3rd person," and to use a mouse as the main character. Road to VR's review, played seated, found that the game's fixed perspective "makes for a perfectly comfortable game."

In 2021 Oculus published a VR locomotion design guide and noted that, although most of it addresses first-person applications, "many of the techniques will be useful in 3rd person applications as well."

Research

A 2006 paper by Patrick Salamin, Daniel Thalmann and Frédéric Vexo at the ACM Symposium on Virtual Reality Software and Technology (VRST) argued that not being able to see one's own body in a VR simulation is sometimes disturbing, and proposed letting users choose between first- and third-person perspectives as in video games such as GTA. Because gamers tend to prefer third person for movement and first person for fine operations, the authors set out to test whether the same holds in VR and augmented reality simulations.

In a 2017 study in Frontiers in Robotics and AI, Geoffrey Gorisse and colleagues had 28 participants deflect projectiles and navigate between platforms in an immersive environment, with the third-person camera placed 3 meters behind the avatar and rotating around the character in proportion to the participant's head rotation. First person scored significantly higher for self-location and ownership of the virtual body; agency showed no significant difference, and spatial presence was high in both conditions. Participants in third person perceived incoming projectiles sooner, while first person was faster for precise hand interactions and for navigation.

A 2017 PLOS ONE study by Henrique Galvan Debarba and colleagues, with 48 participants, found that illusory ownership of a virtual body could be achieved in both first- and third-person perspectives when visual, motor and tactile feedback were congruent. Body ownership and reactions to a virtual threat were generally stronger in first person and in a condition where participants could switch perspective at will than in third person alone, which the authors read as showing that "the possibility of alternating perspective is compatible with a strong sense of embodiment."

Diego Monteiro and colleagues, writing in Computer Animation and Virtual Worlds in 2018, compared 1PP and 3PP on both a head-mounted display and a conventional display using a popular game. They reported that 3PP in VR was less likely to cause simulator sickness than 1PP in VR and was not perceived as immersive, that presence was not required for enjoyment, and that there was no clear preference between the two perspectives for VR gameplay.

Hybrid first- and third-person techniques

Several research techniques use a temporary third-person view for travel while keeping first person for close interaction:

• Outstanding (Sebastian Cmentowski, Andrey Krekhov and Jens Krüger, CHI PLAY 2019) lets players switch on demand from first person to a third-person bird's-eye perspective, in which they are scaled to a giant size above their avatar and set navigation targets for it by raycast aiming. In a between-subjects study with 30 participants against arc-based teleportation, it significantly improved spatial orientation, with no significant differences in cybersickness, presence, enjoyment or competence.
• Out-of-body locomotion (Nathan Navarro Griffin and Eelke Folmer, VRST 2019) lets users switch from a first-person to a third-person avatar view. In third person the camera stays still unless the user moves, so no optical flow is generated while the avatar is steered. The view returns to first person when the user breaks line of sight with the avatar or presses a button to rejoin it. In a study with 22 participants on an obstacle course, it needed 67% fewer viewpoint transitions than teleportation with no significant difference in performance, and participants judged it to be faster.

Problems

A third-person camera is not automatically comfortable. Playful's Paul Bettner said the parameters that control camera motion in Lucky's Tale "are very sensitive" and operate "within a very narrow band of comfort." Early in development it was easy to make people sick with the wrong values, and the camera quickly became uncomfortable if it accelerated or decelerated too fast or made sudden movements. Bettner said Playful's early prototypes showed that action felt best within about arm's length and stopped feeling as good once it was 10 to 15 feet away.

Gunfire Games reported losing sight of the character when it left one room and entered the next in Herobound, physical discomfort when a player must follow a character while looking up or down for too long, and the risk of the character being hidden between floors on ladders. UploadVR's review of Chronos found its locked camera angles "incredibly restrictive" in practice; the reviewer walked off ledges, missed clues and got turned around because of the shifting perspectives. Road to VR's Astro Bot Rescue Mission review noted that players occasionally have to "crane your neck straight up, down, or look directly behind you." Studies also report a weaker sense of body ownership in third person than in first person, and Monteiro and colleagues found 3PP in VR was not perceived as immersive.

Solutions

Designers have used several approaches to keep third-person VR comfortable:

• Restricting camera motion. Playful kept the Lucky's Tale camera within a narrow comfort band and solved problems through level design instead; the team avoided backtracking and added teleporters to bring players back to points of decision.
• Changing the world's scale. Playful "shrank everything by increasing the IPD," making the player a giant and Lucky tiny, so the gameplay area sits about 1 to 10 feet from the player.
• Anchoring the character on camera cuts. In Chronos, each new camera shows the character in the same screen-space position as the previous camera, so the character acts as an anchor point. Cuts happen at natural thresholds such as doorways, camera positions are built around a relaxed, centered head position, ladders are pushed away from the camera, and the game takes control of the character on ladders so its position is known when the camera changes.
• Keeping the camera still. Moss uses a static view of each scene, and out-of-body locomotion keeps the camera stationary while the user steers the avatar in third person.