EvidMTL: Evidential Multi-Task Learning for Uncertainty-Aware Semantic Surface Mapping from Monocular RGB Images

2025-03-06

AI Summary

Existing semantic mapping methods in unstructured environments suffer from overconfident semantic predictions, sparse and noisy depth perception, and inconsistent map representations. To address these issues, this paper proposes EvidKimera—a novel framework that, for the first time, integrates evidential uncertainty modeling into the end-to-end semantic surface mapping pipeline. We design an evidential depth loss to jointly optimize depth estimation and semantic segmentation under a principled uncertainty-aware objective. EvidKimera unifies monocular depth estimation, semantic segmentation, multi-task learning, and Kimera SLAM. On NYU Depth V2, it yields well-calibrated uncertainty estimates. With zero-shot transfer to ScanNetV2, EvidKimera achieves significantly higher accuracy and consistency in semantic surface maps compared to the baseline Kimera system.

Technology Category

• Reasoning under Uncertainty: Uncertainty Representations
• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Multi-modal Vision

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

For scene understanding in unstructured environments, an accurate and uncertainty-aware metric-semantic mapping is required to enable informed action selection by autonomous systems.Existing mapping methods often suffer from overconfident semantic predictions, and sparse and noisy depth sensing, leading to inconsistent map representations. In this paper, we therefore introduce EvidMTL, a multi-task learning framework that uses evidential heads for depth estimation and semantic segmentation, enabling uncertainty-aware inference from monocular RGB images. To enable uncertainty-calibrated evidential multi-task learning, we propose a novel evidential depth loss function that jointly optimizes the belief strength of the depth prediction in conjunction with evidential segmentation loss. Building on this, we present EvidKimera, an uncertainty-aware semantic surface mapping framework, which uses evidential depth and semantics prediction for improved 3D metric-semantic consistency. We train and evaluate EvidMTL on the NYUDepthV2 and assess its zero-shot performance on ScanNetV2, demonstrating superior uncertainty estimation compared to conventional approaches while maintaining comparable depth estimation and semantic segmentation. In zero-shot mapping tests on ScanNetV2, EvidKimera outperforms Kimera in semantic surface mapping accuracy and consistency, highlighting the benefits of uncertainty-aware mapping and underscoring its potential for real-world robotic applications.

Problem

• Overconfident semantic predictions in existing mapping methods
• Sparse and noisy depth sensing in unstructured environments
• Inconsistent map representations for autonomous systems

Innovation

• Evidential heads for depth and semantic tasks
• Novel evidential depth loss function
• EvidKimera for improved 3D semantic consistency