Signing Outside the Studio: Benchmarking Background Robustness for Continuous Sign Language Recognition
This work addresses the robustness of CSLR models for real-world applications where signing occurs outside studios, representing an incremental improvement in domain-specific robustness.
The authors tackled the problem of continuous sign language recognition (CSLR) models failing under background shifts by evaluating existing models on a new benchmark with diverse backgrounds and proposing a training scheme with background randomization and feature disentanglement, achieving improved generalization to unseen backgrounds with minimal additional training.
The goal of this work is background-robust continuous sign language recognition. Most existing Continuous Sign Language Recognition (CSLR) benchmarks have fixed backgrounds and are filmed in studios with a static monochromatic background. However, signing is not limited only to studios in the real world. In order to analyze the robustness of CSLR models under background shifts, we first evaluate existing state-of-the-art CSLR models on diverse backgrounds. To synthesize the sign videos with a variety of backgrounds, we propose a pipeline to automatically generate a benchmark dataset utilizing existing CSLR benchmarks. Our newly constructed benchmark dataset consists of diverse scenes to simulate a real-world environment. We observe even the most recent CSLR method cannot recognize glosses well on our new dataset with changed backgrounds. In this regard, we also propose a simple yet effective training scheme including (1) background randomization and (2) feature disentanglement for CSLR models. The experimental results on our dataset demonstrate that our method generalizes well to other unseen background data with minimal additional training images.