Lost in Compression: the Impact of Lossy Image Compression on Variable Size Object Detection within Infrared Imagery

Abstract

Lossy image compression strategies allow for more efficient storage and transmission of data by encoding data to a reduced form. This is essential enable training with larger datasets on less storage-equipped environments. However, such compression can cause severe decline in performance of deep Convolution Neural Network (CNN) architectures even when mild compression is applied and the resulting compressed imagery is visually identical. In this work, we apply the lossy JPEG compression method with six discrete levels of increasing compression {95, 75, 50, 15, 10, 5} to infrared band (thermal) imagery. Our study quantitatively evaluates the affect that increasing levels of lossy compression has upon the performance of characteristically diverse object detection architectures (Cascade-RCNN, FSAF and Deformable DETR) with respect to varying sizes of objects present in the dataset. When training and evaluating on uncompressed data as a baseline, we achieve maximal mean Average Precision (mAP) of 0.823 with Cascade R-CNN across the FLIR dataset, outperforming prior work. The impact of the lossy compression is more extreme at higher compression levels (15, 10, 5) across all three CNN architectures. However, re-training models on lossy compressed imagery notably ameliorated performances for all three CNN models with an average increment of ∼ 76% (at higher compression level 5). Additionally, we demonstrate the relative sensitivity of differing object areas {tiny, small, medium, large} with respect to the compression level. We show that tiny and small objects are more sensitive to compression than medium and large objects. Overall, Cascade R-CNN attains the maximal mAP across most of the object area categories.

1st Workshop on Maritime Computer Vision (MaCVi) 2023: Challenge Results

“…We incorporate a task-specific model initialisation or transfer learning strategy in this challenge. In our prior work of [13], we analysed the impact of object area on the performance of object detectors. As the target dataset (SeaDronesSee v2) primarily consists of aerial imagery (air-to-ground) with small-size object area (object-area < 32 × 32 pixel [13]), we initialised our model with pre-trained on VisDrone [10] (aerial imagery dataset) instead of the commonly used ImageNet [37] or MS-COCO [61] datasets.…”