Pith. sign in

REVIEW 10 cited by

ResizeMix: Mixing Data with Preserved Object Information and True Labels

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2012.11101 v1 pith:E7OHHDWH submitted 2020-12-21 cs.CV

Jie Qin , Jiemin Fang , Qian Zhang , Wenyu Liu , Xingang Wang , Xinggang Wang This is my paper
classification cs.CV
keywords dataimageinformationmixingaugmentationmethodsobjectcutmix
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Data augmentation is a powerful technique to increase the diversity of data, which can effectively improve the generalization ability of neural networks in image recognition tasks. Recent data mixing based augmentation strategies have achieved great success. Especially, CutMix uses a simple but effective method to improve the classifiers by randomly cropping a patch from one image and pasting it on another image. To further promote the performance of CutMix, a series of works explore to use the saliency information of the image to guide the mixing. We systematically study the importance of the saliency information for mixing data, and find that the saliency information is not so necessary for promoting the augmentation performance. Furthermore, we find that the cutting based data mixing methods carry two problems of label misallocation and object information missing, which cannot be resolved simultaneously. We propose a more effective but very easily implemented method, namely ResizeMix. We mix the data by directly resizing the source image to a small patch and paste it on another image. The obtained patch preserves more substantial object information compared with conventional cut-based methods. ResizeMix shows evident advantages over CutMix and the saliency-guided methods on both image classification and object detection tasks without additional computation cost, which even outperforms most costly search-based automatic augmentation methods.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Learning from Synthetic Data via Provenance-Based Input Gradient Guidance

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    A framework that applies provenance-based guidance to input gradients during synthetic data training to promote learning from target regions only.

  2. AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition

    cs.CV 2026-07 accept novelty 6.0 of 10

    AGVBench benchmarks 30 augmentation strategies for vein recognition and finds mixing methods improve accuracy but harm calibration and adversarial robustness.

  3. $S^{2}$-FracMix: Label-Preserving Self-Saliency Mixup Augmentation

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    S²-FracMix creates label-preserving augmented samples via intra-image self-saliency patch mixing and fractal pattern injection, claiming SOTA results across seven benchmarks in classification, robustness, detection, a...

  4. MMTalker: Multiresolution 3D Talking Head Synthesis with Multimodal Feature Fusion

    cs.CV 2026-04 unverdicted novelty 6.0 of 10

    MMTalker combines multi-resolution mesh sampling with residual graph convolutions and dual cross-attention to synthesize accurate 3D talking head motions from audio.

  5. Enhancing Generalization in Data-free Quantization via Mixup-class Prompting

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Using two class labels in text prompts to generate synthetic calibration images improves data-free post-training quantization accuracy, especially in low-bit settings.

  6. InstructMixup: Instruction-Guided Salient Patch Editing for Robust Data Augmentation

    cs.CV 2026-07 conditional novelty 5.0 of 10

    InstructMixup augments images by using a saliency map to select patches, editing them with an offline instruction-guided generative model, adding fractal texture, and blending the result back into the same image, impr...

  7. AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition

    cs.CV 2026-07 conditional novelty 5.0 of 10

    On vein-recognition benchmarks, mixup-style augmentations win on clean accuracy but hurt calibration and adversarial robustness, while simple geometric transforms usually hurt performance.

  8. MMTalker: Multiresolution 3D Talking Head Synthesis with Multimodal Feature Fusion

    cs.CV 2026-04 unverdicted novelty 5.0 of 10

    MMTalker maps speech to 3D face motion via multi-resolution UV sampling and residual-GCN plus dual cross-attention multimodal fusion, reporting better lip and eye sync than prior methods.

  9. Enhancing Ambiguous Dynamic Facial Expression Recognition with Soft Label-based Data Augmentation

    cs.CV 2025-06 conditional novelty 5.0 of 10

    MIDAS, a mixup-style augmentation for soft-labeled video, improves dynamic facial expression recognition accuracy over hard-label training on DFEW and the new FERV39k-Plus dataset.

  10. A Comparison of Data Augmentation Methods for Training Deep Neural Networks on Synthetic Aperture Sonar

    cs.LG 2026-07 conditional novelty 4.0 of 10

    Optimized CV and physics-informed augmentations usually improve SAS-ATR DNNs, multi-aug policies add diminishing gains on Swin, and benefits often do not transfer to ResNet.

Pith tools