Monocular vision-based 3D object detection is crucial in various sectors, yet
existing methods face significant challenges in terms of accuracy and
computational efficiency. Building on the successful strategies in 2D detection
and depth estimation, we propose MonoDETRNext, which seeks to optimally balance
precision and processing speed. Our methodology includes the development of an
efficient hybrid visual encoder, enhancement of depth prediction mechanisms,
and introduction of an innovative query generation strategy, augmented by an
advanced depth predictor. Building on MonoDETR, MonoDETRNext introduces two
variants: MonoDETRNext-F, which emphasizes speed, and MonoDETRNext-A, which
focuses on precision. We posit that MonoDETRNext establishes a new benchmark in
monocular 3D object detection and opens avenues for future research. We
conducted an exhaustive evaluation demonstrating the model's superior
performance against existing solutions. Notably, MonoDETRNext-A demonstrated a
4.60% improvement in the AP3D metric on the KITTI test benchmark over MonoDETR,
while MonoDETRNext-F showed a 2.21% increase. Additionally, the computational
efficiency of MonoDETRNext-F slightly exceeds that of its predecessor.

基于单目视觉的三维物体检测在各个领域都十分重要，然而现有方法在准确性和计算效率方面面临着重大挑战。在 2D 检测和深度估计方面取得成功的策略基础上，我们提出了 MonoDETRNext，旨在实现优化的准确性和处理速度的平衡。我们的方法包括开发高效的混合视觉编码器，增强深度预测机制，并引入创新的查询生成策略，辅以先进的深度预测器。MonoDETRNext 基于 MonoDETR，并引入了两个变体：MonoDETRNext-F 注重速度，MonoDETRNext-A 注重精度。我们认为 MonoDETRNext 建立了单目三维物体检测的新基准，并为未来的研究开辟了新方向。我们进行了详尽的评估，证明了该模型在性能上优于现有解决方案。值得注意的是，MonoDETRNext-A 在 KITTI 测试基准上的 AP3D 指标相比 MonoDETR 提高了 4.60%，而 MonoDETRNext-F 则提高了 2.21%。此外，MonoDETRNext-F 的计算效率略高于其前身。