#masked depth modeling
9 related articles
·3 min
The Depth Perception Challenge for Transparent Objects: How LingBot-Depth Breaks Through with Masked Depth Modeling
Depth perception for transparent and reflective objects has long been a core challenge in robotic grasping. LingBot-Depth uses masked depth modeling to turn sensor failure into supervisory signals, inferring glass depth from RGB context.
Read more →
·3 min
The Depth Perception Challenge for Transparent Objects: How LingBot-Depth Breaks Through with Masked Depth Modeling
Transparent and reflective object depth perception is a core challenge in robotic grasping. LingBot-Depth uses masked depth modeling to turn sensor failure into supervision, inferring glass depth from RGB context.
Read more →
·3 min
Why Do Transformer Learning Paths Always Start Too Late? Order Matters More Than Materials
Most Transformer learning paths jump straight into Attention Is All You Need, skipping the pain points of sequence models—leaving learners reciting terms without real intuition. Here's a better four-stage order.
Read more →
·4 min
A Guide to Avoiding Pitfalls in PCA Visualization of Self-Supervised Vision Models: Why the Colors Are Meaningless
Running self-supervised vision models (SSL) on a MacBook CPU isn't hard. This article reveals the core misconception of PCA visualization through ViT-S experiments: colors can't convey semantics across images, and changing resolution reverses hues entirely.
Read more →
·7 min
Training Machine Learning Models from Scratch: A Complete Beginner's Guide
A complete walkthrough of training machine learning models from scratch—covering problem definition, data preprocessing, algorithm selection, hyperparameter tuning, and evaluation, with tool recommendations for beginners.
Read more →
·6 min
DINOv2 vs SigLIP: Why the Huge Gap in k-NN Classification? A Guide to Avoiding Vision Encoder Selection Pitfalls
Experiments show DINOv2 Giant scores just 41% on k-NN classification, while SigLIP2 reaches 92%. This article dives into the embedding-space differences between contrastive and self-supervised learning to guide vision encoder selection.
Read more →
·7 min
GLM-5.2 Released: Open-Source Coding Capability Approaches Closed-Source Flagships for the First Time
GLM-5.2 tops open-weight models in coding with a 74.4 Frontiers-WE score, beating GPT-5.5. Its MIT license enables local deployment, and the gap with closed-source flagships is closing fast.
Read more →
·4 min
FTPO: A Training-Level Solution to AI Repetition Loops
FTPO (Final Token Preference Optimization) tackles AI "Doom Loops" at the training level rather than patching them at inference time — exploring its principles, value for small/quantized models, and open challenges.
Read more →
Expert Opinions·3 min
The Salary Ceiling for Agent Engineers: Two Critical Dividing Lines
Agent engineer salary gaps hinge on two dividing lines: real production deployment experience and depth of foundational theory including deep learning, fine-tuning, and reinforcement learning.
Read more →