A Composite Dual Attention Transformer for Query-Based Instance Segmentation in Autonomous Driving
Bibliographic record
Abstract
High-precision instance segmentation is a fundamental requirement for applications such as autonomous driving, where accurately identifying objects like pedestrians and vehicles is essential. To address the limitations of existing models in capturing complex features and global context, we propose an enhanced QueryInst framework. Our approach integrates three key architectural advancements. We first employ a Dual Attention Vision Transformer (DaViT) backbone to extract rich, multiscale feature representations. This is further strengthened by a Composite Backbone Network (CBNet) architecture, which harnesses the output of multiple DaViT backbones to construct a more powerful and representative feature hierarchy. A Feature Pyramid Network with CARAFE upsampling is then used to enhance multi-scale feature fusion and recover fine spatial details. Extensive experiments on the Cityscapes dataset demonstrate the superior accuracy of our model, achieving an AP of 38.2% and an <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{AP}_{50}$</tex> of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$65.6 \%$</tex>. These results signify improvements of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$3.8 \%$</tex> and 6% over the baseline, outperforming other state-of-the-art methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".