mlsanityData to 14 Oct 2026

ShapeFormer: Shape Prior Visible-to-Amodal Transformer-based Amodal Instance Segmentation

arXiv 2403.11376 ↗First posted 18 Mar 2024cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Amodal Instance Segmentation (AIS) presents a challenging task as it involves predicting both visible and occluded parts of objects within images. Existing AIS methods rely on a bidirectional approach, encompassing both the transition from amodal features to visible features (amodal-to-visible) and from visible features to amodal features (visible-to-amodal). Our observation shows that the utilization of amodal features through the amodal-to-visible can confuse the visible features due to the extra information of occluded/hidden segments not presented in visible display. Consequently, this compromised quality of visible features during the subsequent visible-to-amodal transition. To tackle this issue, we introduce ShapeFormer, a decoupled Transformer-based model with a visible-to-amodal transition. It facilitates the explicit relationship between output segmentations and avoids the need for amodal-to-visible transitions. ShapeFormer comprises three key modules: (i) Visible-Occluding Mask Head for predicting visible segmentation with occlusion awareness, (ii) Shape-Prior Amodal Mask Head for predicting amodal and occluded masks, and (iii) Category-Specific Shape Prior Retriever aims to provide shape prior knowledge. Comprehensive experiments and extensive ablation studies across various AIS benchmarks demonstrate the effectiveness of our ShapeFormer. The code is available at: urlhttps://github.com/UARK-AICV/ShapeFormer

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 21 of its 25 references that are in the map. First posted on arXiv 18 Mar 2024.

Builds on

21 of 25 references.

References · 21

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2017Mask R-CNNcited by 7,277
  6. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,577
  7. 2016An overview of gradient descent optimization algorithmscited by 1,033
  8. 2021Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayerscited by 104
  9. 2015Semantic Amodal Segmentationcited by 103
  10. 2020Self-Supervised Scene De-occlusioncited by 96
  11. 2017RegNet: Multimodal Sensor Registration Using Deep Neural Networkscited by 69
  12. 2021Mask Transfiner for High-Quality Instance Segmentationcited by 64
9 more
  1. 2016Amodal Instance Segmentationcited by 63
  2. 2018Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentationcited by 62
  3. 2020Amodal Segmentation Based on Visible Region Segmentation and Shape Priorcited by 43
  4. 2021Unseen Object Amodal Instance Segmentation via Hierarchical Occlusion Modelingcited by 43
  5. 2022Amodal Panoptic Segmentationcited by 38
  6. 2022AISFormer: Amodal Instance Segmentation with Transformercited by 35
  7. 2021A Weakly Supervised Amodal Segmenter with Boundary Uncertainty Estimationcited by 21
  8. 2022Self-supervised Amodal Video Object Segmentationcited by 14
  9. 2020Learning Vector Quantized Shape Code for Amodal Blastomere Instance Segmentationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 1; 2026: 3.

6 papers.

  1. 2024Amodal Instance Segmentation with Diffusion Shape Prior Estimationcited by 5
  2. 2026Amodal SAM: A Unified Amodal Segmentation Framework with Generalizationcited by 2
  3. 2025ViTA-Seg: Vision Transformer for Amodal Segmentation in Roboticscited by 1
  4. 2026Learnable Shape Prototypes with Occlusion-Geometry-Guided Injection for Amodal Instance Segmentationcited by 1
  5. 2024A2VIS: Amodal-Aware Approach to Video Instance Segmentationcited by 0
  6. 2026QCell: Recombining and Aligning Cell Queries for Overlapping Instance Segmentationcited by 0