Abstract We introduce Spatial-Temporal Memory Networks for video object detection. At its
core, a novel Spatial-Temporal Memory module (STMM) serves as the recurrent …

Recent years have witnessed a trend of applying context frames to boost the performance of
object detection as video object detection. Existing methods usually aggregate features at …

Extending state-of-the-art object detectors from image to video is challenging. The accuracy
of detection suffers from degenerated object appearances in videos, eg, motion blur, video …

Single-frame object detectors perform well on videos sometimes, even without temporal
context. However, challenges such as occlusion, motion blur, and rare poses of objects are …

State-of-the-art video object detection methods maintain a memory structure, either a sliding
window or a memory queue, to enhance the current frame using attention mechanisms …

Recently, DETR and Deformable DETR have been proposed to eliminate the need for many
hand-designed components in object detection while demonstrating good performance as …

The existing methods for video object detection mainly depend on two-stage image object
detectors. The fact that two-stage detectors are generally slow makes it difficult to apply in …

Transferring image-based object detectors to the domain of videos remains a challenging
problem. Previous efforts mostly exploit optical flow to propagate features across frames …

Video object detection is more challenging than image object detection because of the
deteriorated frame quality. To enhance the feature representation, state-of-the-art methods …

Transferring existing image-based detectors to the video is non-trivial since the quality of
frames is always deteriorated by part occlusion, rare pose, and motion blur. Previous …