Temporal Feature Networks for CNN based Object Detection
This work addresses the need for temporal understanding in object detection for autonomous driving, but it is incremental as it builds on existing CNN architectures.
The paper tackled the problem of incorporating temporal information into CNN-based object detectors to improve environment perception, achieving competitive results on the KITTI dataset compared to a non-temporal baseline.
For reliable environment perception, the use of temporal information is essential in some situations. Especially for object detection, sometimes a situation can only be understood in the right perspective through temporal information. Since image-based object detectors are currently based almost exclusively on CNN architectures, an extension of their feature extraction with temporal features seems promising. Within this work we investigate different architectural components for a CNN-based temporal information extraction. We present a Temporal Feature Network which is based on the insights gained from our architectural investigations. This network is trained from scratch without any ImageNet information based pre-training as these images are not available with temporal information. The object detector based on this network is evaluated against the non-temporal counterpart as baseline and achieves competitive results in an evaluation on the KITTI object detection dataset.