# Flattening the Parent Bias: Hierarchical Semantic Segmentation in the Poincar'e Ball arXiv: https://arxiv.org/abs/2404.03778 ยท HTML: https://mlsanity.com/paper/2404.03778/ ## Key facts - Cited by 9 papers in the map (data through 14 Oct 2026). - Builds on 40 of its 64 references that are in the map. - First posted on arXiv 4 Apr 2024. ## Builds on (most cited first, 30 of 40) - [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015) - [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017) - [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014) - [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016) - [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018) - [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013) - [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018) - [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018) - [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://mlsanity.com/paper/2107.06278/) (2021) - [Poincar'e Embeddings for Learning Hierarchical Representations](https://mlsanity.com/paper/1705.08039/) (2017) - [Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations](https://mlsanity.com/paper/1707.03237/) (2017) - [Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation](https://mlsanity.com/paper/1909.11065/) (2019) - [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://mlsanity.com/paper/2202.11094/) (2022) - [Hyperbolic Neural Networks](https://mlsanity.com/paper/1805.09112/) (2018) - [Semi-Supervised Semantic Segmentation with Cross-Consistency Training](https://mlsanity.com/paper/2003.09005/) (2020) - [Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration](https://mlsanity.com/paper/1910.12656/) (2019) - [Representation Tradeoffs for Hyperbolic Embeddings](https://mlsanity.com/paper/1804.03329/) (2018) - [Boundary IoU: Improving Object-Centric Image Segmentation Evaluation](https://mlsanity.com/paper/2103.16562/) (2021) - [IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments](https://mlsanity.com/paper/1811.10200/) (2018) - [Semi-Supervised Semantic Segmentation with High- and Low-level Consistency](https://mlsanity.com/paper/1908.05724/) (2019) - [Making Better Mistakes: Leveraging Class Hierarchies with Deep Networks](https://mlsanity.com/paper/1912.09393/) (2019) - [Hyperbolic Vision Transformers: Combining Improvements in Metric Learning](https://mlsanity.com/paper/2203.10833/) (2022) - [Hyperbolic Image Segmentation](https://mlsanity.com/paper/2203.05898/) (2022) - [Deep Hierarchical Semantic Segmentation](https://mlsanity.com/paper/2203.14335/) (2022) - [Semantic Segmentation using Adversarial Networks](https://mlsanity.com/paper/1611.08408/) (2016) - [Hyperbolic Deep Learning in Computer Vision: A Survey](https://mlsanity.com/paper/2305.06611/) (2023) - [Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifiers](https://mlsanity.com/paper/2107.11472/) (2021) - [Learning the Predictability of the Future](https://mlsanity.com/paper/2101.01600/) (2021) - [Dynamic-structured Semantic Propagation Network](https://mlsanity.com/paper/1803.06067/) (2018) - [Hierarchical Human Parsing with Typed Part-Relation Reasoning](https://mlsanity.com/paper/2003.04845/) (2020) ## Built on by (most cited first, 9 of 9) - [Hyperbolic Category Discovery](https://mlsanity.com/paper/2504.06120/) (2025) - [Taxonomy-Aware Continual Semantic Segmentation in Hyperbolic Spaces for Open-World Perception](https://mlsanity.com/paper/2407.18145/) (2024) - [Tree-Wasserstein Distance for High Dimensional Data with a Latent Feature Hierarchy](https://mlsanity.com/paper/2410.21107/) (2024) - [BlurDM: A Blur Diffusion Model for Image Deblurring](https://mlsanity.com/paper/2512.03979/) (2025) - [Metric Convolutions: A Unifying Theory to Adaptive Image Convolutions](https://mlsanity.com/paper/2406.05400/) (2024) - [Adversarial Attacks on Hyperbolic Networks](https://mlsanity.com/paper/2412.01495/) (2024) - [Segment Any Vehicle: Semantic and Visual Context Driven SAM and A Benchmark](https://mlsanity.com/paper/2508.04260/) (2025) - [HNDiff: Haze-Noise Diffusion for Image Dehazing](https://mlsanity.com/paper/2608.10995/) (2026) - [Lorentz Framework for Semantic Segmentation](https://mlsanity.com/paper/2604.16836/) (2026) ## Abstract (arXiv) Hierarchy is a natural representation of semantic taxonomies, including the ones routinely used in image segmentation. Indeed, recent work on semantic segmentation reports improved accuracy from supervised training leveraging hierarchical label structures. Encouraged by these results, we revisit the fundamental assumptions behind that work. We postulate and then empirically verify that the reasons for the observed improvement in segmentation accuracy may be entirely unrelated to the use of the semantic hierarchy. To demonstrate this, we design a range of cross-domain experiments with a representative hierarchical approach. We find that on the new testing domains, a flat (non-hierarchical) segmentation network, in which the parents are inferred from the children, has superior segmentation accuracy to the hierarchical approach across the board. Complementing these findings and inspired by the intrinsic properties of hyperbolic spaces, we study a more principled approach to hierarchical segmentation using the Poincar'e ball model. The hyperbolic representation largely outperforms the previous (Euclidean) hierarchical approach as well and is on par with our flat Euclidean baseline in terms of segmentation accuracy. However, it additionally exhibits surprisingly strong calibration quality of the parent nodes in the semantic hierarchy, especially on the more challenging domains. Our combined analysis suggests that the established practice of hierarchical segmentation may be limited to in-domain settings, whereas flat classifiers generalize substantially better, especially if they are modeled in the hyperbolic space.