无需重训练，用梯度法评估单目深度估计的不确定性。
Revisiting Gradient-based Uncertainty for Monocular Depth Estimation
- 通过特征图梯度计算不确定性，不依赖真实深度标签。
- 在KITTI和NYU数据集上有效识别高风险预测区域。
- 适合自动驾驶等对可靠性要求高的场景使用。
单目深度估计易受图像模糊性（如动态物体、阴影）影响，导致错误预测。为保障安全关键应用，需对每个像素进行不确定性评估。本文提出一种后处理方法，基于梯度的不确定性估计，适用于已训练好的深度模型。通过引入基于预测深度与参考深度一致性的辅助损失函数，生成伪真实深度作为参考，该参考深度由简单的图像或特征增强获得，方法简洁高效。利用反向传播计算单层或多层特征图相对于深度预测的导数，得到最终不确定性分数。实验表明，该方法在标准基准KITTI和NYU上无需重新训练即可有效识别不确定区域，尤其对仅用单目序列训练的模型表现更优。代码与模型已公开：https://github.com/jhornauer/GrUMoDepth。
原文摘要 · Abstract (English)
Monocular depth estimation, similar to other image-based tasks, is prone to erroneous predictions due to ambiguities in the image, for example, caused by dynamic objects or shadows. For this reason, pixel-wise uncertainty assessment is required for safety-critical applications to highlight the areas where the prediction is unreliable. We address this in a post hoc manner and introduce gradient-based uncertainty estimation for already trained depth estimation models. To extract gradients without depending on the ground truth depth, we introduce an auxiliary loss function based on the consistency of the predicted depth and a reference depth. The reference depth, which acts as pseudo ground truth, is in fact generated using a simple image or feature augmentation, making our approach simple and effective. To obtain the final uncertainty score, the derivatives w.r.t. the feature maps from single or multiple layers are calculated using back-propagation. We demonstrate that our gradient-based approach is effective in determining the uncertainty without re-training using the two standard depth estimation benchmarks KITTI and NYU. In particular, for models trained with monocular sequences and therefore most prone to uncertainty, our method outperforms related approaches. In addition, we publicly provide our code and models: https://github.com/jhornauer/GrUMoDepth
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。