LightDepth: A Resource Efficient Depth Estimation Approach for Dealing with Ground Truth Sparsity via Curriculum Learning
This work addresses computational constraints for robots and drones, offering an incremental improvement in efficiency.
The paper tackles the problem of resource-intensive depth estimation for autonomous devices by proposing a fast and battery-efficient approach, achieving a 71% improvement in response time while maintaining accuracy comparable to state-of-the-art models.
Advances in neural networks enable tackling complex computer vision tasks such as depth estimation of outdoor scenes at unprecedented accuracy. Promising research has been done on depth estimation. However, current efforts are computationally resource-intensive and do not consider the resource constraints of autonomous devices, such as robots and drones. In this work, we present a fast and battery-efficient approach for depth estimation. Our approach devises model-agnostic curriculum-based learning for depth estimation. Our experiments show that the accuracy of our model performs on par with the state-of-the-art models, while its response time outperforms other models by 71%. All codes are available online at https://github.com/fatemehkarimii/LightDepth.