88Z AI, 8 July — Ant Lingbo Technology has released the spatial perception model LingBot-Depth 2.0, with training data significantly expanded to 150 million. It performed excellently in depth completion benchmarks. The company also open-sourced the visual foundation model LingBot-Vision and achieved certification cooperation with Orbbec.
At a glance
- LingBot-Depth 2.0 training data expanded to 150 million, halving indoor depth error
- Open-sourced LingBot-Vision achieves sub-pixel boundary localization and gains Orbbec certification
The update
Ant Lingbo Technology has officially launched the spatial perception model LingBot-Depth 2.0. The model's training dataset has been significantly expanded from the original 3 million to 150 million entries. In 16 benchmark evaluations related to depth completion, the model secured first place in 12 categories and reduced indoor depth error by half. At the same time, the company open-sourced the visual foundation model LingBot-Vision. This model pioneers boundary structure pre-training technology in the industry, achieving sub-pixel level boundary localization using only 160 million images. Currently, LingBot-Depth 2.0 has passed Orbbec certification, and both parties plan to collaborate on launching related SDK development kits and integrated camera products.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 蚂蚁灵波科技Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.