mlsanityData to 7 Oct 2026

A Partial Replication of MaskFormer in TensorFlow on TPUs for the TensorFlow Model Garden

arXiv 2404.18801 ↗First posted 29 Apr 2024cs.CV cs.LG cs.SEcited by 1

Not on the map. Why?

Abstract (arXiv)

This paper undertakes the task of replicating the MaskFormer model a universal image segmentation model originally developed using the PyTorch framework, within the TensorFlow ecosystem, specifically optimized for execution on Tensor Processing Units (TPUs). Our implementation exploits the modular constructs available within the TensorFlow Model Garden (TFMG), encompassing elements such as the data loader, training orchestrator, and various architectural components, tailored and adapted to meet the specifications of the MaskFormer model. We address key challenges encountered during the replication, non-convergence issues, slow training, adaptation of loss functions, and the integration of TPU-specific functionalities. We verify our reproduced implementation and present qualitative results on the COCO dataset. Although our implementation meets some of the objectives for end-to-end reproducibility, we encountered challenges in replicating the PyTorch version of MaskFormer in TensorFlow. This replication process is not straightforward and requires substantial engineering efforts. Specifically, it necessitates the customization of various components within the TFMG, alongside thorough verification and hyper-parameter tuning. The replication is available at: https://github.com/PurdueDualityLab/tf-maskformer/tree/main/official/projects/maskformer

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 20 of its 25 references that are in the map. First posted on arXiv 29 Apr 2024.

Builds on

20 of 25 references.

References · 20

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  5. 2023Segment Anythingcited by 8,158
  6. 2020End-to-End Object Detection with Transformerscited by 7,384
  7. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
  8. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  9. 2015SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentationcited by 1,483
  10. 2021Per-Pixel Classification is Not All You Need for Semantic Segmentationcited by 1,148
  11. 2018Panoptic Segmentationcited by 796
  12. 2019UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentationcited by 583
8 more
  1. 2021Pre-Trained Models: Past, Present and Futurecited by 319
  2. 2020Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)cited by 318
  3. 2023An Empirical Study of Pre-Trained Model Reuse in the Hugging Face Deep Learning Model Registrycited by 45
  4. 2023Challenges and Practices of Deep Learning Model Reengineering: A Case Study on Computer Visioncited by 11
  5. 2024Reusing Deep Learning Models: Challenges and Directions in Software Engineeringcited by 10
  6. 2022An Empirical Study of Challenges in Converting Deep Learning Modelscited by 8
  7. 2021An Experience Report on Machine Learning Reproducibility: Guidance for Practitioners and TensorFlow Model Garden Contributorscited by 6
  8. 2023Discrepancies among Pre-trained Deep Neural Networks: A New Threat to Model Zoo Reliabilitycited by 4

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Improving the Reproducibility of Deep Learning Software: An Initial Investigation through a Case Study Analysiscited by 3