AI Paper Atlas

Research map / Point clouds / Point cloud classification

TOD3Cap: Towards 3D Dense Captioning in Outdoor Scenes

Bu Jin, Yupeng Zheng, Pengfei Li, Weize Li, Yuhang Zheng, Sujie Hu, Xinyu Liu, Jinwei Zhu, Zhijie Yan, Haiyang Sun, Kun Zhan, Peng Jia and 3 more

ECCV 2024 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 34 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Relation Q-Former instead of relational graph (ablation)TOD3Cap network · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 82.7 -> 94.1Improved
use Relation Q-Former instead of transformer decoder (ablation)TOD3Cap network · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 90.0 -> 94.1Improved
use LLaMA-Adapter decoder instead of GPT2 (ablation)TOD3Cap network · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 85.6 -> 94.1Improved
use LLaMA instead of S&T decoder (ablation)TOD3Cap network · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 78.6 -> 94.1Improved
add detector pretraining stage (ablation)TOD3Cap network training · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 69.5 -> 94.1Improved
add captioner pretraining stage (ablation)TOD3Cap network training · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 85.3 -> 94.1Improved
add camera images to LiDAR-only input (ablation)TOD3Cap network · TOD3Cap valCIDEr@0.5IoU: 74.4 -> 108.0Improved
add LiDAR to camera-only input (ablation)TOD3Cap network · TOD3Cap valCIDEr@0.5IoU: 94.1 -> 108.0Improved
use larger BEV resolution and backbone (Small to full) (ablation)TOD3Cap network · TOD3Cap val (2D camera input)CIDEr@0.5IoU: 87.5 -> 94.1Improved
adapt indoor baselines with the BEV detectorScan2Cap · TOD3Captraining: training collapse -> C@0.5 62.5 (2D+3D)Improved
add LLM-generated driving instructions as input to UniADUniAD · nuScenes planningAvg. collision: 0.31 -> 0.23Improved
add LLM-generated driving instructions as input to UniADUniAD · nuScenes planningAvg. L2: 1.03 -> 1.05Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy