# Towards Enhancing Fine-grained Details for Image Matting

arXiv: https://arxiv.org/abs/2101.09095 · HTML: https://mlsanity.com/paper/2101.09095/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 38 references that are in the map.
- First posted on arXiv 22 Jan 2021.

## Builds on (most cited first, 14 of 14)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation](https://mlsanity.com/paper/1511.00561/) (2015)
- [Deep Image Matting](https://mlsanity.com/paper/1703.03872/) (2017)
- [Boundary-Aware Feature Propagation for Scene Segmentation](https://mlsanity.com/paper/1909.00179/) (2019)
- [Indices Matter: Learning to Index for Deep Image Matting](https://mlsanity.com/paper/1908.00672/) (2019)
- [Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation](https://mlsanity.com/paper/1909.09725/) (2019)
- [Semantic Correlation Promoted Shape-Variant Context for Segmentation](https://mlsanity.com/paper/1909.02651/) (2019)
- [Disentangled Image Matting](https://mlsanity.com/paper/1909.04686/) (2019)
- [Bi-directional Dermoscopic Feature Learning and Multi-scale Consistent Decision Fusion for Skin Lesion Segmentation](https://mlsanity.com/paper/2002.08694/) (2020)

## Built on by (most cited first, 12 of 12)

- [MatteFormer: Transformer-Based Image Matting via Prior-Tokens](https://mlsanity.com/paper/2203.15662/) (2022)
- [Instance-Specific Feature Propagation for Referring Segmentation](https://mlsanity.com/paper/2204.12109/) (2022)
- [Knowledge-aware Deep Framework for Collaborative Skin Lesion Segmentation and Melanoma Recognition](https://mlsanity.com/paper/2106.03455/) (2021)
- [Self-Regularized Prototypical Network for Few-Shot Semantic Segmentation](https://mlsanity.com/paper/2210.16829/) (2022)
- [Deep Image Matting: A Comprehensive Survey](https://mlsanity.com/paper/2304.04672/) (2023)
- [Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives](https://mlsanity.com/paper/2205.12627/) (2022)
- [dugMatting: Decomposed-Uncertainty-Guided Matting](https://mlsanity.com/paper/2306.01452/) (2023)
- [Alpha Matte Generation from Single Input for Portrait Matting](https://mlsanity.com/paper/2106.03210/) (2021)
- [Spatial Feature Mapping for 6DoF Object Pose Estimation](https://mlsanity.com/paper/2206.01831/) (2022)
- [Towards Label-Efficient Human Matting: A Simple Baseline for Weakly Semi-Supervised Trimap-Free Human Matting](https://mlsanity.com/paper/2404.00921/) (2024)
- [DiffusionMat: Alpha Matting as Sequential Refinement Learning](https://mlsanity.com/paper/2311.13535/) (2023)
- [SAM2Matting: Generalized Image and Video Matting](https://mlsanity.com/paper/2606.27339/) (2026)

## Abstract (arXiv)

In recent years, deep natural image matting has been rapidly evolved by extracting high-level contextual features into the model. However, most current methods still have difficulties with handling tiny details, like hairs or furs. In this paper, we argue that recovering these microscopic details relies on low-level but high-definition texture features. However, {these features are downsampled in a very early stage in current encoder-decoder-based models, resulting in the loss of microscopic details}. To address this issue, we design a deep image matting model {to enhance fine-grained details. Our model consists of} two parallel paths: a conventional encoder-decoder Semantic Path and an independent downsampling-free Textural Compensate Path (TCP). The TCP is proposed to extract fine-grained details such as lines and edges in the original image size, which greatly enhances the fineness of prediction. Meanwhile, to leverage the benefits of high-level context, we propose a feature fusion unit(FFU) to fuse multi-scale features from the semantic path and inject them into the TCP. In addition, we have observed that poorly annotated trimaps severely affect the performance of the model. Thus we further propose a novel term in loss function and a trimap generation method to improve our model's robustness to the trimaps. The experiments show that our method outperforms previous start-of-the-art methods on the Composition-1k dataset.