NDC-Scene: Boost Monocular 3D Semantic Scene Completion in Normalized Device Coordinates Space
This work improves 3D scene understanding from single images for applications like autonomous driving and robotics, though it is incremental as it builds on prior SSC methods.
The paper tackles monocular 3D semantic scene completion by addressing feature ambiguity, pose ambiguity, and computation imbalance in existing methods, proposing NDC-Scene which extends 2D features to normalized device coordinates space and uses a depth-adaptive dual decoder, resulting in consistent outperformance of state-of-the-art methods on SemanticKITTI and NYUv2 datasets.
Monocular 3D Semantic Scene Completion (SSC) has garnered significant attention in recent years due to its potential to predict complex semantics and geometry shapes from a single image, requiring no 3D inputs. In this paper, we identify several critical issues in current state-of-the-art methods, including the Feature Ambiguity of projected 2D features in the ray to the 3D space, the Pose Ambiguity of the 3D convolution, and the Computation Imbalance in the 3D convolution across different depth levels. To address these problems, we devise a novel Normalized Device Coordinates scene completion network (NDC-Scene) that directly extends the 2D feature map to a Normalized Device Coordinates (NDC) space, rather than to the world space directly, through progressive restoration of the dimension of depth with deconvolution operations. Experiment results demonstrate that transferring the majority of computation from the target 3D space to the proposed normalized device coordinates space benefits monocular SSC tasks. Additionally, we design a Depth-Adaptive Dual Decoder to simultaneously upsample and fuse the 2D and 3D feature maps, further improving overall performance. Our extensive experiments confirm that the proposed method consistently outperforms state-of-the-art methods on both outdoor SemanticKITTI and indoor NYUv2 datasets. Our code are available at https://github.com/Jiawei-Yao0812/NDCScene.