mlsanityData to 14 Oct 2026

THIRDEYE: Cue-Aware Monocular Depth Estimation via Brain-Inspired Multi-Stage Fusion

arXiv 2506.20877 ↗First posted 25 Jun 2025cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Monocular depth estimation methods traditionally train deep models to infer depth directly from RGB pixels. This implicit learning often overlooks explicit monocular cues that the human visual system relies on, such as occlusion boundaries, shading, and perspective. Rather than expecting a network to discover these cues unaided, we present ThirdEye, a cue-aware pipeline that deliberately supplies each cue through specialised, pre-trained, and frozen networks. These cues are fused in a three-stage cortical hierarchy (V1->V2->V3) equipped with a key-value working-memory module that weights them by reliability. An adaptive-bins transformer head then produces a high-resolution disparity map. Because the cue experts are frozen, ThirdEye inherits large amounts of external supervision while requiring only modest fine-tuning. This extended version provides additional architectural detail, neuroscientific motivation, and an expanded experimental protocol; quantitative results will appear in a future revision.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 63 references that are in the map. First posted on arXiv 25 Jun 2025.

Builds on

22 of 63 references.

References · 22

  1. 2016Densely Connected Convolutional Networkscited by 6,720
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  3. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  4. 2021Vision Transformers for Dense Predictioncited by 1,803
  5. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  6. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 737
  7. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 657
  8. 2014Neural Turing Machinescited by 644
  9. 2020AdaBins: Depth Estimation using Adaptive Binscited by 601
  10. 2015Holistically-Nested Edge Detectioncited by 592
  11. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 506
  12. 2014Recurrent Models of Visual Attentioncited by 349
10 more
  1. 2016Deep Predictive Coding Networks for Video Prediction and Unsupervised Learningcited by 204
  2. 2020On the uncertainty of self-supervised monocular depth estimationcited by 182
  3. 2019HorizonNet: Learning Room Layout with 1D Representation and Pano Stretch Data Augmentationcited by 108
  4. 2021Vision Permutator: A Permutable MLP-Like Architecture for Visual Recognitioncited by 103
  5. 2019Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNscited by 90
  6. 2016One-shot Learning with Memory-Augmented Neural Networkscited by 60
  7. 2019Self-calibrating Deep Photometric Stereo Networkscited by 47
  8. 2025One-shot Entropy Minimizationcited by 32
  9. 2016Depth from a Single Image by Harmonizing Overcomplete Local Network Predictionscited by 16
  10. 2024Uncertainty-guided annotation enhances segmentation with the human-in-the-loopcited by 4

Built on by

Not cited yet.