MelNet: A Real-Time Deep Learning Algorithm for Object Detection

TL;DR

MelNet is a new deep learning model for object detection that, after training on the KITTI dataset, achieves competitive accuracy comparable to pre-trained models, demonstrating the effectiveness of transfer learning and scenario-specific training.

Contribution

Introduces MelNet, a novel deep learning algorithm for object detection, and compares its performance with existing models, highlighting the benefits of training on specific datasets.

Findings

MelNet achieved an mAP of 0.732 after 300 epochs.

Training MelNet on KITTI surpasses EfficientDet after 150 epochs.

Transfer learning with pre-trained models yields superior results.

Abstract

In this study, a novel deep learning algorithm for object detection, named MelNet, was introduced. MelNet underwent training utilizing the KITTI dataset for object detection. Following 300 training epochs, MelNet attained an mAP (mean average precision) score of 0.732. Additionally, three alternative models -YOLOv5, EfficientDet, and Faster-RCNN-MobileNetv3- were trained on the KITTI dataset and juxtaposed with MelNet for object detection. The outcomes underscore the efficacy of employing transfer learning in certain instances. Notably, preexisting models trained on prominent datasets (e.g., ImageNet, COCO, and Pascal VOC) yield superior results. Another finding underscores the viability of creating a new model tailored to a specific scenario and training it on a specific dataset. This investigation demonstrates that training MelNet exclusively on the KITTI dataset also surpasses EfficientDet after 150 epochs. Consequently, post-training, MelNet's performance closely aligns with that of other pre-trained models.