Forget Less, Count Better: A Domain-Incremental Self-Distillation Learning Benchmark for Lifelong Crowd Counting. (arXiv:2205.03307v1 [cs.CV])

Web: http://arxiv.org/abs/2205.03307

Crowd Counting has important applications in public safety and pandemic
control. A robust and practical crowd counting system has to be capable of
continuously learning with the new-coming domain data in real-world scenarios
instead of fitting one domain only. Off-the-shelf methods have some drawbacks
to handle multiple domains. 1) The models will achieve limited performance
(even drop dramatically) among old domains after training images from new
domains due to the discrepancies of intrinsic data distributions from various
domains, which is …

arxiv benchmark count cv distillation incremental learning