Research map / Detection and segmentation / CNN features
StageInteractor: Query-based Object Detector with Cross-stage Interaction
Yao Teng, Haisong Liu, Sheng Guo, Limin Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add cross-stage label assignment (ablation) | AdaMixer ResNet-50 100 queries · COCO minival 1x | AP: 43.0 -> 44.8 | Improved |
| add cross-stage dynamic filter reuse (ablation) | AdaMixer ResNet-50 100 queries · COCO minival 1x | AP: 44.1 -> 44.8 | Improved |
| use cross-stage label assigner instead of hybrid layer matching (ablation) | AdaMixer 100 queries · COCO minival 1x | AP: 41.8 -> 44.8 | Improved |
| add IoU score to cross-stage label gathering (ablation) | cross-stage label assigner · COCO minival 1x | AP: 43.1 -> 44.8 | Improved |
| gather labels from all previous stages (alpha_i=1) instead of from i-1 (ablation) | cross-stage label assigner · COCO minival 1x | AP: 44.8 -> 43.3 | Hurt |
| reuse more spatial filters (3 instead of 1) (ablation) | spatial dynamic mixing · COCO minival 1x | AP: 44.8 -> 44.4 | Hurt |
| increase number of reused channel-mixing filters (ablation) | channel mixing · COCO minival 1x | AP: 44.1 -> 44.8 | Improved |
| fuse previous and current filters with adapters instead of using one directly (ablation) | channel mixing · COCO minival 1x | AP: 44.5 -> 44.8 | Improved |
| also modify localization supervision in the assigner (ablation) | cross-stage label assigner · COCO minival 1x | AP: 44.8 -> 44.8 | No clear effect |
| add cross-stage label assignment to DINO | DINO ResNet-50 12 epochs · COCO minival | AP: 49.2 -> 49.7 | Improved |
| add cross-stage label assignment to DINO with Swin-B | DINO Swin-B · COCO minival | AP: 55.8 -> 56.2 | Improved |
| add cross-stage label assignment and filter reuse | AdaMixer · COCO minival speed | inference speed (img/s): 13.5 -> 11.6 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AdaMixer — AdaMixer: A Fast-Converging Query-Based Object Detector (CVPR 2022) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Enhanced Training of Query-Based Object Detection via Selective Query Recollection (CVPR 2023 · similarity 0.88)
- Deep Equilibrium Object Detection and Segmentation (TPAMI 2025 · similarity 0.88)
- Order Matters: Competition-Guided Query Ordering for RNN-Based Object Detection (NeurIPS 2026 · similarity 0.85)
- CQ-DINO: Mitigating Gradient Dilution via Category Queries for Vast Vocabulary Object Detection (NeurIPS 2025 · similarity 0.84)
- DetKDS: Knowledge Distillation Search for Object Detectors (ICML 2024 · similarity 0.84)
- EASE-DETR: Easing the Competition among Object Queries (CVPR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
