Pokemon Go的公司Niantic用收集的数据创建空间情报模型

痴货

Pokemon Go这类强空间交互的游戏收集了大量的图片，数据，以及地理信息用来训练AI模型，并且这类模型能够提供甚至到厘米级别的精度。然而其制作公司和CIA有千丝万缕的联系。

1. https://nianticlabs.com/news/largegeospatialmodel?hl=en

复制代码

Over the past five years, Niantic has focused on building our Visual Positioning System (**), which uses a single image from a phone to determine its position and orientation using a 3D map built from people scanning interesting locations in our games and Scaniverse.

With **, users can position themselves in the world with centimeter-level accuracy. That means they can see digital content placed against the physical environment precisely and realistically. This content is persistent in that it stays in a location after you’ve left, and it’s then shareable with others. For example, we recently started rolling out an experimental feature in Pokémon GO, called Pokémon Playgrounds, where the user can place Pokémon at a specific location, and they will remain there for others to see and interact with.

Niantic’s ** is built from user scans, taken from different perspectives and at various times of day, at many times during the years, and with positioning information attached, creating a highly detailed understanding of the world. This data is unique because it is taken from a pedestrian perspective and includes places inaccessible to cars.

Today we have 10 million scanned locations around the world, and over 1 million of those are activated and available for use with our ** service. We receive about 1 million fresh scans each week, each containing hundreds of discrete images.

As part of the **, we build classical 3D vision maps using structure from motion techniques - but also a new type of neural map for each place. These neural models, based on our research papers ACE (2023) and ACE Zero (2024) do not represent locations using classical 3D data structures anymore, but encode them implicitly in the learnable parameters of a neural network. These networks can swiftly compress thousands of mapping images into a lean, neural representation. Given a new query image, they offer precise positioning for that location with centimeter-level accuracy.

这个公司被CIA的白手套投资公司In-Q-Tel投资过，而且此公司的CEO John Hanke本身就和CIA关系匪浅：

1. https://lunduke.locals.com/post/5756204/the-cia-nsa-and-pok-mon-go

复制代码

Way back in 2001, Keyhole, Inc. was founded by John Hanke (who previously worked in a “foreign affairs” position within the U.S. government). The company was named after the old “eye-in-the-sky” military satellites. One of the key, early backers of Keyhole was a firm called In-Q-Tel.

In-Q-Tel is the venture capital firm of the CIA. Yes, the Central Intelligence Agency. Much of the funding purportedly came from the National Geospatial-Intelligence Agency (NGA). The NGA handles combat support for the U.S. Department of Defense and provides intelligence to the NSA and CIA, among others.

Keyhole’s noteworthy public product was “Earth.” Renamed to “Google Earth” after Google acquired Keyhole in 2004.

In 2010, Niantic Labs was founded (inside Google) by Keyhole’s founder, John Hanke.

Over the next few years, Niantic created two location-based apps/games. The first was Field Trip, a smartphone application where users walk around and find things. The second was Ingress, a sci-fi-themed game where players walk around and between locations in the real world.

In 2015, Niantic was spun off from Google and became its own company. Then Pokémon Go was developed and launched by Niantic. It’s a game where you walk around in the real world (between locations suggested by the service) while holding your smartphone.

The original CEO of In-Q-Tel was a man named Gilman Louie. Louie received multiple awards for his work with In-Q-Tel - including CIA Agency Seal Medallions, Director's Award by the Director of the Central Intelligence Agency, and Director of National Intelligence Medallion – which included investing in Keyhole.

Louie now sits on the board of directors of Niantic.

kenqqex

ingress本来就是间谍游戏

Vtuber

间谍游戏石锤了

痴货

但是从技术角度分析这个的确很有用处，大模型在根据图片查询推理地理位置上大有可为。

补充一下他们的论文，这个技术挺有意思：

1. https://nianticlabs.github.io/mickey/

复制代码

cosx

高瞻远瞩

Viteeee

现成的数据放在那，傻逼才不用——任何有AI产品公司都是这么想的。