Robot Vision Achieves New Breakthrough! Ant Group's LingBot-Depth 2.0 Spatial Perception Model Officially Released

On July 7, Lingbo Technology—an embodied intelligence company under Ant Group—released the spatial perception model LingBot-Depth 2.0. Trained on a dataset of 150 million samples, the model achieves comprehensive upgrades in edge clarity, fine object recognition, long-range depth estimation, and robustness in complex scenarios. LingBot-Depth is a spatial perception model independently developed by Lingbo. It acts as the "eyes" of robots in the physical world. The first version solved the challenge of spatial perception in complex scenarios such as transparent and reflective surfaces. Compared to LingBot-Depth 1.0, the training data for version 2.0 has expanded from 3 million to 150 million samples, leading to overall performance improvements: it achieved first place in 12 out of 16 evaluation items in depth completion benchmarks. In the most challenging scenario—large-scale depth missing indoors—the depth error was halved compared to the previous generation (RMSE dropped from 0.132 to 0.062). It performed particularly well in scenarios where traditional depth cameras often fail, such as with glass, mirrors, and transparent objects. At the same time, LingBot-Depth 2.0 introduced its visual foundation model—LingBot-Vision—building a capability chain from "understanding" to "accurate perception." This integration aims to address core challenges in robot vision, including spatial perception, fine recognition, and complex environment adaptation.

(Figure 1: LingBot-Depth 2.0 completes a complete and flat 3D structure in difficult scenarios like mirrors and glass)

The breakthrough of LingBot-Depth 2.0 benefits from the outstanding visual representation capabilities of LingBot-Vision. As a general-purpose visual model, LingBot-Vision is the first visual foundation model in the industry to use "boundary structure" as a pre-training objective, achieving a breakthrough in the training paradigm for spatial perception. It possesses sub-pixel-level boundary positioning and spatial structure understanding capabilities, delivering higher precision and more stable spatial perception. The pre-training corpus of LingBot-Vision consists of only 160 million images—one order of magnitude smaller than DINOv3—yet its depth estimation accuracy surpasses that of DINOv3. Moreover, LingBot-Vision’s judgment of object boundaries is sufficiently stable to enable continuous tracking of object boundaries in videos. This release of LingBot-Vision includes four versions: ViT-G/L/B/S. According to available information, LingBot-Vision not only supports the training of LingBot-Depth 2.0 but also has versatile applications in multiple domains.

(Figure 2: LingBot-Depth 2.0 performs well in real sensor depth completion tests)

(Figure 3: Compared to mainstream visual foundation models, LingBot-Vision identifies object boundaries and spatial structures more clearly and stably)

Currently, LingBot-Depth 2.0 has passed the professional certification of Orbbec Depth Vision Lab. Practical scenario tests show that, based on the chip-level 3D raw data provided by Orbbec's Gemini330 series stereo 3D camera, LingBot-Depth 2.0 has significantly improved edge clarity, object contour integrity, fine object recognition, long-range depth estimation, and robustness in complex lighting and material scenarios.

(Figure 4: LingBot-Depth 2.0 passed the professional evaluation of Orbbec Depth Vision Lab, demonstrating extremely high accuracy and stability in spatial and temporal depth estimation tasks across multiple types of sensors)

In terms of commercialization, Ant Lingbo has engaged in in-depth cooperation with Orbbec across many areas. According to the information, the latest RGB-D version of the EGO device in Orbbec’s new no-body data acquisition product matrix will be adapted to the LingBot-Depth version optimized by Lingbo for data acquisition scenarios. In the future, this collaboration will further integrate higher-level commercial versions, continuously fill depth gaps, optimize object edges and spatial structure details, and provide more accurate, stable, and usable real-world data for embodied intelligence model training. Additionally, Orbbec will launch an SDK product integrating the latest model capabilities of LingBot-Depth, available for use by robot customers at the edge, allowing robots using the Gemini330 series camera to achieve better depth effects. Orbbec also plans to launch an integrated camera product with the commercial version of LingBot-Depth by the end of the year, realizing the integration of "3D camera + spatial perception capabilities." With the release of these two models, their collaboration is expected to expand into more fields.