CVAug 15, 2023

Identity-Consistent Aggregation for Video Object Detection

arXiv:2308.07737v111 citationsh-index: 31Has Code
Originality Highly original
AI Analysis

This work addresses inefficiencies in video object detection for applications requiring real-time processing, though it is incremental as it builds on existing VID models with a novel aggregation method.

The paper tackles the problem of video object detection by focusing on identity-consistent temporal contexts to handle appearance variations like occlusion and motion blur, achieving state-of-the-art performance with 84.7% mAP on ImageNet VID and running 7x faster at 39.3 fps.

In Video Object Detection (VID), a common practice is to leverage the rich temporal contexts from the video to enhance the object representations in each frame. Existing methods treat the temporal contexts obtained from different objects indiscriminately and ignore their different identities. While intuitively, aggregating local views of the same object in different frames may facilitate a better understanding of the object. Thus, in this paper, we aim to enable the model to focus on the identity-consistent temporal contexts of each object to obtain more comprehensive object representations and handle the rapid object appearance variations such as occlusion, motion blur, etc. However, realizing this goal on top of existing VID models faces low-efficiency problems due to their redundant region proposals and nonparallel frame-wise prediction manner. To aid this, we propose ClipVID, a VID model equipped with Identity-Consistent Aggregation (ICA) layers specifically designed for mining fine-grained and identity-consistent temporal contexts. It effectively reduces the redundancies through the set prediction strategy, making the ICA layers very efficient and further allowing us to design an architecture that makes parallel clip-wise predictions for the whole video clip. Extensive experimental results demonstrate the superiority of our method: a state-of-the-art (SOTA) performance (84.7% mAP) on the ImageNet VID dataset while running at a speed about 7x faster (39.3 fps) than previous SOTAs.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes