Chat-3D/Chat-3D-v2
Code for "Chat-3D v2: Bridging 3D Scene and Large Language Models with Object Identifiers"
PythonMIT
Issues
- 1
The error during traing stage.
#36 opened by hanxunyu - 2
the relation module for v2.1 version
#35 opened by PanXiongAdam - 4
About the epoches in Training&Inference.
#34 opened by KaKa-101 - 19
About data preprocessing.
#31 opened by KaKa-101 - 1
Save test output
#32 opened by yitianlian - 3
Training datasets for v2.1 version
#33 opened by ZCMax - 5
- 3
The question about preprocessing dataset
#22 opened by hanxunyu - 5
About the annotation files
#23 opened by KaKa-101 - 2
about parameters
#19 opened by iris0329 - 2
ScanQA finetuning
#13 opened by cy94 - 9
- 1
Using llama3
#29 opened by WeitaiKang - 4
Point cloud data feature
#28 opened by KimWu1994 - 7
Ask for extracting input for 3D encoder
#25 opened by jkstyle2 - 17
Running Inference on ScanRefer
#26 opened by jkstyle2 - 1
About stage 3
#20 opened by Chunmian-art - 2
object id
#16 opened by kuaileqipaoshui - 3
about table 3
#14 opened by iris0329 - 2
ModuleNotFoundError: No module named 'data' ModuleNotFoundError: No module named 'models'
#17 opened by Chunmian-art - 8
- 2
ImportError: FlashAttention2 has been toggled on
#21 opened by hanxunyu - 2
Why each object's color could be represented as a 3x4 channel tensor in training stage 1
#18 opened by iris0329 - 3
Load vicuna weights
#10 opened by WeitaiKang - 2
About the Identifier-rich Scene Captioning Dataset
#15 opened by ZCMax - 5
Color of each instance
#9 opened by LilRedWu - 2
Ask for visualization
#12 opened by WeitaiKang - 2
Ask for PointGroup results
#11 opened by WeitaiKang - 7
Prepare annotation of private dataset
#8 opened by WeitaiKang - 2
evaluation
#7 opened by kuaileqipaoshui - 1
About trainned wights
#4 opened by Hank10032 - 4
- 4
stage3 traning
#5 opened by kuaileqipaoshui - 1
Data Generation
#3 opened by Joy-R - 1
- 2
CMT module
#2 opened by DevXiaoYan