Research map / Detection and segmentation / Open-set and few-shot detection
MonoTTA: Fully Test-Time Adaptation for Monocular 3D Object Detection
Hongbin Lin, Yifan Zhang, Shuaicheng Niu, Shuguang Cui, Zhen Li
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 reliability-driven adaptation loss L_AO (ablation) | MonoGround · KITTI-C level 1 (avg) | mean AP3D|R40: 5.17 -> 14.40 | Improved |
| add negative regularization L_Nreg (ablation) | MonoGround · KITTI-C level 1 (avg) | mean AP3D|R40: 5.17 -> 12.15 | Improved |
| combine L_AO with L_Nreg (ablation) | MonoGround · KITTI-C level 1 (avg) | mean AP3D|R40: 14.40 -> 16.79 | Improved |
| combine L_AO with L_Nreg at severity level 2 (ablation) | MonoGround · KITTI-C level 2 (avg) | mean AP3D|R40: 10.96 -> 12.85 | Improved |
| use MonoTTA instead of TENT on MonoGround cars | MonoGround · KITTI-C level 1 Car | mean AP3D|R40: 27.54 -> 32.64 | Improved |
| use MonoTTA instead of EATA on MonoGround cars | MonoGround · KITTI-C level 1 Car | mean AP3D|R40: 25.28 -> 32.64 | Improved |
| apply any TTA method at higher severity (level 2) | TTA methods on KITTI-C · KITTI-C level 2 | mean AP3D|R40 | Mixed |
| add MonoTTA to an instance-level (batch size 1) detector | MonoNeRD · KITTI-C level 1 | mean AP3D|R40: 9.36 -> 17.38 | Improved |
| use MonoTTA instead of TENT on MonoNeRD cyclists | MonoNeRD · KITTI-C level 1 Cyclist | mean AP3D|R40: 4.75 -> 4.35 | Hurt |
| use entropy minimization on all detections (TENT) in the real N->D task | TENT · nuScenes Night to Daytime | mAP | Hurt |
Later papers that compared with this method
2 later accepted papers report MonoTTA as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts (ICCV 2025 · better on 10 of 10 shared results)
- Self-Bootstrapping for Versatile Test-Time Adaptation (ICML 2025 · better on 3 of 3 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
- Unsupervised Domain Adaptation for Monocular 3D Object Detection via Self-Training (ECCV 2022 · similarity 0.90)
- MonoTAKD: Teaching Assistant Knowledge Distillation for Monocular 3D Object Detection (CVPR 2025 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.88)
- MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models (CVPR 2024 · similarity 0.88)
- MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders (NeurIPS 2024 · similarity 0.87)
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.87)
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.
