# SNIPER: Efficient Multi-Scale Training

arXiv: https://arxiv.org/abs/1805.09300 · HTML: https://mlsanity.com/paper/1805.09300/

## Key facts

- Cited by 88 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 May 2018.

## Built on by (most cited first, 30 of 88)

- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis](https://mlsanity.com/paper/2111.14791/) (2021)
- [Embracing Single Stride 3D Object Detector with Sparse Transformer](https://mlsanity.com/paper/2112.06375/) (2021)
- [Dynamic Head: Unifying Object Detection Heads with Attentions](https://mlsanity.com/paper/2106.08322/) (2021)
- [RangeDet:In Defense of Range View for LiDAR-based 3D Object Detection](https://mlsanity.com/paper/2103.10039/) (2021)
- [Towards Large-Scale Small Object Detection: Survey and Benchmarks](https://mlsanity.com/paper/2207.14096/) (2022)
- [You Only Look One-level Feature](https://mlsanity.com/paper/2103.09460/) (2021)
- [QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection](https://mlsanity.com/paper/2103.09136/) (2021)
- [Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection](https://mlsanity.com/paper/2103.17115/) (2021)
- [A Normalized Gaussian Wasserstein Distance for Tiny Object Detection](https://mlsanity.com/paper/2110.13389/) (2021)
- [RFLA: Gaussian Receptive Field based Label Assignment for Tiny Object Detection](https://mlsanity.com/paper/2208.08738/) (2022)
- [Dynamic Coarse-to-Fine Learning for Oriented Tiny Object Detection](https://mlsanity.com/paper/2304.08876/) (2023)
- [Tracking Pedestrian Heads in Dense Crowd](https://mlsanity.com/paper/2103.13516/) (2021)
- [ASM-Loc: Action-aware Segment Modeling for Weakly-Supervised Temporal Action Localization](https://mlsanity.com/paper/2203.15187/) (2022)
- [Graph-DETR3D: Rethinking Overlapping Regions for Multi-View 3D Object Detection](https://mlsanity.com/paper/2204.11582/) (2022)
- [UFPMP-Det: Toward Accurate and Efficient Object Detection on Drone Imagery](https://mlsanity.com/paper/2112.10415/) (2021)
- [No More Strided Convolutions or Pooling: A New CNN Building Block for Low-Resolution Images and Small Objects](https://mlsanity.com/paper/2208.03641/) (2022)
- [Multi-Scale Aligned Distillation for Low-Resolution Detection](https://mlsanity.com/paper/2109.06875/) (2021)
- [Scale-aware Automatic Augmentation for Object Detection](https://mlsanity.com/paper/2103.17220/) (2021)
- [Image Data Augmentation Approaches: A Comprehensive Survey and Future directions](https://mlsanity.com/paper/2301.02830/) (2023)
- [End-to-End Spatio-Temporal Action Localisation with Video Transformers](https://mlsanity.com/paper/2304.12160/) (2023)
- [GiraffeDet: A Heavy-Neck Paradigm for Object Detection](https://mlsanity.com/paper/2202.04256/) (2022)
- [Cascaded Zoom-in Detector for High Resolution Aerial Images](https://mlsanity.com/paper/2303.08747/) (2023)
- [Scale-Equivalent Distillation for Semi-Supervised Object Detection](https://mlsanity.com/paper/2203.12244/) (2022)
- [ESOD: Efficient Small Object Detection on High-Resolution Images](https://mlsanity.com/paper/2407.16424/) (2024)
- [SparseDet: Improving Sparsely Annotated Object Detection with Pseudo-positive Mining](https://mlsanity.com/paper/2201.04620/) (2022)
- [Open World Object Detection: A Survey](https://mlsanity.com/paper/2410.11301/) (2024)
- [YOLO-Z: Improving small object detection in YOLOv5 for autonomous vehicles](https://mlsanity.com/paper/2112.11798/) (2021)
- [You Should Look at All Objects](https://mlsanity.com/paper/2207.07889/) (2022)
- [GenMM: Geometrically and Temporally Consistent Multimodal Data Generation for Video and LiDAR](https://mlsanity.com/paper/2406.10722/) (2024)

## Abstract (arXiv)

We present SNIPER, an algorithm for performing efficient multi-scale training in instance level visual recognition tasks. Instead of processing every pixel in an image pyramid, SNIPER processes context regions around ground-truth instances (referred to as chips) at the appropriate scale. For background sampling, these context-regions are generated using proposals extracted from a region proposal network trained with a short learning schedule. Hence, the number of chips generated per image during training adaptively changes based on the scene complexity. SNIPER only processes 30% more pixels compared to the commonly used single scale training at 800x1333 pixels on the COCO dataset. But, it also observes samples from extreme resolutions of the image pyramid, like 1400x2000 pixels. As SNIPER operates on resampled low resolution chips (512x512 pixels), it can have a batch size as large as 20 on a single GPU even with a ResNet-101 backbone. Therefore it can benefit from batch-normalization during training without the need for synchronizing batch-normalization statistics across GPUs. SNIPER brings training of instance level recognition tasks like object detection closer to the protocol for image classification and suggests that the commonly accepted guideline that it is important to train on high resolution images for instance level visual recognition tasks might not be correct. Our implementation based on Faster-RCNN with a ResNet-101 backbone obtains an mAP of 47.6% on the COCO dataset for bounding box detection and can process 5 images per second during inference with a single GPU. Code is available at https://github.com/MahyarNajibi/SNIPER/.