A Density-Guided Temporal Attention Transformer for Indiscernible Object Counting in Underwater Video
This work addresses the problem of counting objects that blend with their surroundings in underwater environments, which is incremental as it builds on existing dense object counting methods.
The paper tackles indiscernible object counting in underwater video by introducing a new large-scale dataset, YoutubeFish-35, and proposing TransVidCount, a method that achieves state-of-the-art performance on this dataset.
Dense object counting or crowd counting has come a long way thanks to the recent development in the vision community. However, indiscernible object counting, which aims to count the number of targets that are blended with respect to their surroundings, has been a challenge. Image-based object counting datasets have been the mainstream of the current publicly available datasets. Therefore, we propose a large-scale dataset called YoutubeFish-35, which contains a total of 35 sequences of high-definition videos with high frame-per-second and more than 150,000 annotated center points across a selected variety of scenes. For benchmarking purposes, we select three mainstream methods for dense object counting and carefully evaluate them on the newly collected dataset. We propose TransVidCount, a new strong baseline that combines density and regression branches along the temporal domain in a unified framework and can effectively tackle indiscernible object counting with state-of-the-art performance on YoutubeFish-35 dataset.