ST-DETR: Spatio-Temporal Object Traces Attention Detection Transformer
This work addresses moving object detection for autonomous driving or video analysis, but it appears incremental as it builds on existing transformer methods with temporal extensions.
The paper tackled object detection from temporal frames by proposing ST-DETR, a spatio-temporal transformer architecture, and achieved a 5% mAP improvement on the KITTI MOD dataset over a spatial baseline.
We propose ST-DETR, a Spatio-Temporal Transformer-based architecture for object detection from a sequence of temporal frames. We treat the temporal frames as sequences in both space and time and employ the full attention mechanisms to take advantage of the features correlations over both dimensions. This treatment enables us to deal with frames sequence as temporal object features traces over every location in the space. We explore two possible approaches; the early spatial features aggregation over the temporal dimension, and the late temporal aggregation of object query spatial features. Moreover, we propose a novel Temporal Positional Embedding technique to encode the time sequence information. To evaluate our approach, we choose the Moving Object Detection (MOD)task, since it is a perfect candidate to showcase the importance of the temporal dimension. Results show a significant 5% mAP improvement on the KITTI MOD dataset over the 1-step spatial baseline.