# What Does the Encoder Actually Decide? A Controlled Comparison of Vision Backbones on Joint Tree Segmentation and Stereo Depth

arXiv: https://arxiv.org/abs/2609.13232 · HTML: https://mlsanity.com/paper/2609.13232/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 30 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model](https://mlsanity.com/paper/2401.09417/) (2024)
- [CoAtNet: Marrying Convolution and Attention for All Data Sizes](https://mlsanity.com/paper/2106.04803/) (2021)
- [MetaFormer Is Actually What You Need for Vision](https://mlsanity.com/paper/2111.11418/) (2021)
- [RepVGG: Making VGG-style ConvNets Great Again](https://mlsanity.com/paper/2101.03697/) (2021)
- [Pyramid Stereo Matching Network](https://mlsanity.com/paper/1803.08669/) (2018)
- [Cross-stitch Networks for Multi-task Learning](https://mlsanity.com/paper/1604.03539/) (2016)
- [End-to-End Learning of Geometry and Context for Deep Stereo Regression](https://mlsanity.com/paper/1703.04309/) (2017)
- [Patches Are All You Need?](https://mlsanity.com/paper/2201.09792/) (2022)
- [An autonomous robot for pruning modern, planar fruit trees](https://mlsanity.com/paper/2206.07201/) (2022)

## Abstract (arXiv)

A robot pruning trees needs two facts per pixel: whether it belongs to a tree, and its distance. Both are usually obtained via task heads attached to a vision backbone chosen by reputation rather than measurement. Holding dataset, decoders, losses, schedule, and evaluation fixed, we ask: how much does the encoder choice change joint semantic segmentation and stereo depth on thin vegetation? We build a hard parameter-sharing network with one encoder feeding both branches, swapping only the encoder without downstream retuning. We evaluate [N] encoders across [M] architecture families (CNNs, transformers, hybrids, MLP-mixers, state-space models) near a ~25M budget, trained from scratch. Depth is evaluated on tree pixels only; segmentation uses boundary F1 and background IoU to prevent "label-everything-tree" shortcuts. Three findings stand out. First, the strongest encoders are convolutional and hybrid, not transformers: [BestEncoder] leads with [MIoU] segmentation mIoU and [Delta] depth δ1, while [X] of [Y] plain vision transformers collapse when trained from scratch. Second, parameter count does not predict quality --- [SmallEncoder] at only [P]M parameters outranks models two orders of magnitude larger. Third, segmentation and depth rankings agree strongly (Spearman ρ = [RhoValue]), showing no task conflict. Finally, [K] of [N] encoders collapse to degenerate all-tree segmentation --- exposed by boundary F1 but hidden by region IoU.