Release MapQR + MapGR nuScenes 24ep checkpoint

README.md

@@ -0,0 +1,56 @@
+---
+license: mit
+tags:
+  - autonomous-driving
+  - hd-map
+  - vectorized-map
+  - nuscenes
+  - mmdetection3d
+datasets:
+  - nuscenes
+---
+
+# MapGR: MapQR + MapGR checkpoint (nuScenes, ResNet-50, 24 epochs)
+
+This repository hosts the released weights for **Learning Global Representation from Queries for Vectorized HD Map Construction** (ICML 2026, [arXiv:2510.06969](https://arxiv.org/abs/2510.06969), [project page](https://xr-lee.github.io/mapgr-demo/)).
+
+Training and evaluation code, installation and data preparation are in the GitHub repository (`mapqr/` sub-project).
+
+| File | Description |
+|---|---|
+| `mapqr_mapgr_nusc_r50_24ep.pth` | Model weights: `state_dict` plus minimal `meta`, no optimizer state. |
+| `mapqr_mapgr_nusc_r50_24ep_config.py` | Fully resolved config, for reference. Train and evaluate with `mapqr/projects/configs/mapgr/mapqr_mapgr_nusc_r50_24ep.py` from the code repository. |
+
+## Results (nuScenes val)
+
+| | mAP<sub>1</sub> | mAP<sub>2</sub> | AP<sub>div</sub> / AP<sub>ped</sub> / AP<sub>bou</sub> (mAP<sub>2</sub> thresholds) |
+|---|---|---|---|
+| Paper (MapQR + MapGR, 24 ep) | 45.3 | 68.1 | 69.8 / 65.8 / 68.8 |
+| This checkpoint | 44.87 | 67.96 | 69.35 / 64.95 / 69.57 |
+
+mAP<sub>2</sub> averages Chamfer AP over the {0.5, 1.0, 1.5} m thresholds; mAP<sub>1</sub> averages over {0.2, 0.5, 1.0} m.
+
+This checkpoint was **re-trained with the released code**. It is not the run reported in the paper.
+
+- Training: 8 GPUs × 4 samples, fp16.
+- The numbers above come from evaluating this exact file with the released `tools/test.py`.
+- Repeated runs of this setting vary by roughly ±0.4 mAP<sub>2</sub>.
+
+```shell
+# inside mapqr/ of the code repository
+wget -P ckpts https://huggingface.co/Lee1413/MapGR/resolve/main/mapqr_mapgr_nusc_r50_24ep.pth
+./tools/dist_test_map.sh projects/configs/mapgr/mapqr_mapgr_nusc_r50_24ep.py ckpts/mapqr_mapgr_nusc_r50_24ep.pth 1
+```
+
+SHA-256 of `mapqr_mapgr_nusc_r50_24ep.pth`: `d4f868312234806b5c0a97ea2c7edf71173de3fac1d510742a99317e1c6f6916`
+
+## Citation
+
+```bibtex
+@inproceedings{qiu2026mapgr,
+  title     = {Learning Global Representation from Queries for Vectorized {HD} Map Construction},
+  author    = {Qiu, Shoumeng and Li, Xinrun and Long, Yang and Xue, Xiangyang and Ojha, Varun and Pu, Jian},
+  booktitle = {International Conference on Machine Learning (ICML)},
+  year      = {2026}
+}
+```

mapqr_mapgr_nusc_r50_24ep.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f868312234806b5c0a97ea2c7edf71173de3fac1d510742a99317e1c6f6916
+size 658832805

mapqr_mapgr_nusc_r50_24ep_config.py

@@ -0,0 +1,485 @@
+point_cloud_range = [-15.0, -30.0, -2.0, 15.0, 30.0, 2.0]
+class_names = [
+    'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier',
+    'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'
+]
+dataset_type = 'CustomNuScenesOfflineLocalMapDataset'
+data_root = 'data/nuscenes/'
+input_modality = dict(
+    use_lidar=False,
+    use_camera=True,
+    use_radar=False,
+    use_map=False,
+    use_external=True)
+file_client_args = dict(backend='disk')
+train_pipeline = [
+    dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+    dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+    dict(type='PhotoMetricDistortionMultiViewImage'),
+    dict(
+        type='NormalizeMultiviewImage',
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375],
+        to_rgb=True),
+    dict(
+        type='LoadPointsFromFile',
+        coord_type='LIDAR',
+        load_dim=5,
+        use_dim=5,
+        file_client_args=dict(backend='disk')),
+    dict(
+        type='CustomPointToMultiViewDepth',
+        downsample=1,
+        grid_config=dict(
+            x=[-30.0, -30.0, 0.15],
+            y=[-15.0, -15.0, 0.15],
+            z=[-10, 10, 20],
+            depth=[1.0, 35.0, 0.5])),
+    dict(type='PadMultiViewImageDepth', size_divisor=32),
+    dict(
+        type='DefaultFormatBundle3D',
+        with_gt=False,
+        with_label=False,
+        class_names=['divider', 'ped_crossing', 'boundary']),
+    dict(type='CustomCollect3D', keys=['img', 'gt_depth'])
+]
+test_pipeline = [
+    dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+    dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+    dict(
+        type='NormalizeMultiviewImage',
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375],
+        to_rgb=True),
+    dict(
+        type='MultiScaleFlipAug3D',
+        img_scale=(1600, 900),
+        pts_scale_ratio=1,
+        flip=False,
+        transforms=[
+            dict(type='PadMultiViewImage', size_divisor=32),
+            dict(
+                type='DefaultFormatBundle3D',
+                with_gt=False,
+                with_label=False,
+                class_names=['divider', 'ped_crossing', 'boundary']),
+            dict(type='CustomCollect3D', keys=['img'])
+        ])
+]
+eval_pipeline = [
+    dict(
+        type='LoadPointsFromFile',
+        coord_type='LIDAR',
+        load_dim=5,
+        use_dim=5,
+        file_client_args=dict(backend='disk')),
+    dict(
+        type='LoadPointsFromMultiSweeps',
+        sweeps_num=10,
+        file_client_args=dict(backend='disk')),
+    dict(
+        type='DefaultFormatBundle3D',
+        class_names=[
+            'car', 'truck', 'trailer', 'bus', 'construction_vehicle',
+            'bicycle', 'motorcycle', 'pedestrian', 'traffic_cone', 'barrier'
+        ],
+        with_label=False),
+    dict(type='Collect3D', keys=['points'])
+]
+data = dict(
+    samples_per_gpu=4,
+    workers_per_gpu=8,
+    train=dict(
+        type='CustomNuScenesOfflineLocalMapDataset',
+        data_root='data/nuscenes/',
+        ann_file='data/nuscenes/nuscenes_map_infos_temporal_train.pkl',
+        pipeline=[
+            dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+            dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+            dict(type='PhotoMetricDistortionMultiViewImage'),
+            dict(
+                type='NormalizeMultiviewImage',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(
+                type='LoadPointsFromFile',
+                coord_type='LIDAR',
+                load_dim=5,
+                use_dim=5,
+                file_client_args=dict(backend='disk')),
+            dict(
+                type='CustomPointToMultiViewDepth',
+                downsample=1,
+                grid_config=dict(
+                    x=[-30.0, -30.0, 0.15],
+                    y=[-15.0, -15.0, 0.15],
+                    z=[-10, 10, 20],
+                    depth=[1.0, 35.0, 0.5])),
+            dict(type='PadMultiViewImageDepth', size_divisor=32),
+            dict(
+                type='DefaultFormatBundle3D',
+                with_gt=False,
+                with_label=False,
+                class_names=['divider', 'ped_crossing', 'boundary']),
+            dict(type='CustomCollect3D', keys=['img', 'gt_depth'])
+        ],
+        classes=[
+            'car', 'truck', 'construction_vehicle', 'bus', 'trailer',
+            'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'
+        ],
+        modality=dict(
+            use_lidar=False,
+            use_camera=True,
+            use_radar=False,
+            use_map=False,
+            use_external=True),
+        test_mode=False,
+        box_type_3d='LiDAR',
+        aux_seg=dict(
+            use_aux_seg=True,
+            bev_seg=True,
+            pv_seg=True,
+            seg_classes=3,
+            feat_down_sample=32,
+            pv_thickness=1),
+        use_valid_flag=True,
+        bev_size=(200, 100),
+        pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+        fixed_ptsnum_per_line=20,
+        eval_use_same_gt_sample_num_flag=True,
+        padding_value=-10000,
+        map_classes=['divider', 'ped_crossing', 'boundary'],
+        queue_length=1),
+    val=dict(
+        type='CustomNuScenesOfflineLocalMapDataset',
+        ann_file='data/nuscenes/nuscenes_map_infos_temporal_val.pkl',
+        pipeline=[
+            dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+            dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+            dict(
+                type='NormalizeMultiviewImage',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(
+                type='MultiScaleFlipAug3D',
+                img_scale=(1600, 900),
+                pts_scale_ratio=1,
+                flip=False,
+                transforms=[
+                    dict(type='PadMultiViewImage', size_divisor=32),
+                    dict(
+                        type='DefaultFormatBundle3D',
+                        with_gt=False,
+                        with_label=False,
+                        class_names=['divider', 'ped_crossing', 'boundary']),
+                    dict(type='CustomCollect3D', keys=['img'])
+                ])
+        ],
+        classes=[
+            'car', 'truck', 'construction_vehicle', 'bus', 'trailer',
+            'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'
+        ],
+        modality=dict(
+            use_lidar=False,
+            use_camera=True,
+            use_radar=False,
+            use_map=False,
+            use_external=True),
+        test_mode=True,
+        box_type_3d='LiDAR',
+        data_root='data/nuscenes/',
+        map_ann_file='data/nuscenes/nuscenes_map_anns_val.json',
+        bev_size=(200, 100),
+        pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+        fixed_ptsnum_per_line=20,
+        eval_use_same_gt_sample_num_flag=True,
+        padding_value=-10000,
+        map_classes=['divider', 'ped_crossing', 'boundary'],
+        samples_per_gpu=1),
+    test=dict(
+        type='CustomNuScenesOfflineLocalMapDataset',
+        data_root='data/nuscenes/',
+        ann_file='data/nuscenes/nuscenes_map_infos_temporal_val.pkl',
+        pipeline=[
+            dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+            dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+            dict(
+                type='NormalizeMultiviewImage',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(
+                type='MultiScaleFlipAug3D',
+                img_scale=(1600, 900),
+                pts_scale_ratio=1,
+                flip=False,
+                transforms=[
+                    dict(type='PadMultiViewImage', size_divisor=32),
+                    dict(
+                        type='DefaultFormatBundle3D',
+                        with_gt=False,
+                        with_label=False,
+                        class_names=['divider', 'ped_crossing', 'boundary']),
+                    dict(type='CustomCollect3D', keys=['img'])
+                ])
+        ],
+        classes=[
+            'car', 'truck', 'construction_vehicle', 'bus', 'trailer',
+            'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'
+        ],
+        modality=dict(
+            use_lidar=False,
+            use_camera=True,
+            use_radar=False,
+            use_map=False,
+            use_external=True),
+        test_mode=True,
+        box_type_3d='LiDAR',
+        map_ann_file='data/nuscenes/nuscenes_map_anns_val.json',
+        bev_size=(200, 100),
+        pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+        fixed_ptsnum_per_line=20,
+        eval_use_same_gt_sample_num_flag=True,
+        padding_value=-10000,
+        map_classes=['divider', 'ped_crossing', 'boundary']),
+    shuffler_sampler=dict(type='DistributedGroupSampler'),
+    nonshuffler_sampler=dict(type='DistributedSampler'))
+evaluation = dict(
+    interval=12,
+    pipeline=[
+        dict(type='LoadMultiViewImageFromFiles', to_float32=True),
+        dict(type='RandomScaleImageMultiViewImage', scales=[0.5]),
+        dict(
+            type='NormalizeMultiviewImage',
+            mean=[123.675, 116.28, 103.53],
+            std=[58.395, 57.12, 57.375],
+            to_rgb=True),
+        dict(
+            type='MultiScaleFlipAug3D',
+            img_scale=(1600, 900),
+            pts_scale_ratio=1,
+            flip=False,
+            transforms=[
+                dict(type='PadMultiViewImage', size_divisor=32),
+                dict(
+                    type='DefaultFormatBundle3D',
+                    with_gt=False,
+                    with_label=False,
+                    class_names=['divider', 'ped_crossing', 'boundary']),
+                dict(type='CustomCollect3D', keys=['img'])
+            ])
+    ],
+    metric='chamfer',
+    save_best='NuscMap_chamfer/mAP',
+    rule='greater')
+checkpoint_config = dict(interval=1, max_keep_ckpts=1)
+log_config = dict(
+    interval=50,
+    hooks=[dict(type='TextLoggerHook'),
+           dict(type='TensorboardLoggerHook')])
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+work_dir = None
+load_from = None
+resume_from = None
+workflow = [('train', 1)]
+plugin = True
+plugin_dir = 'projects/mmdet3d_plugin/'
+voxel_size = [0.15, 0.15, 4.0]
+dbound = [1.0, 35.0, 0.5]
+grid_config = dict(
+    x=[-30.0, -30.0, 0.15],
+    y=[-15.0, -15.0, 0.15],
+    z=[-10, 10, 20],
+    depth=[1.0, 35.0, 0.5])
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+map_classes = ['divider', 'ped_crossing', 'boundary']
+num_vec = 50
+fixed_ptsnum_per_gt_line = 20
+fixed_ptsnum_per_pred_line = 20
+eval_use_same_gt_sample_num_flag = True
+num_map_classes = 3
+_dim_ = 256
+_pos_dim_ = 128
+_ffn_dim_ = 512
+_num_levels_ = 1
+_num_points_in_pillar_ = 8
+bev_h_ = 200
+bev_w_ = 100
+queue_length = 1
+aux_seg_cfg = dict(
+    use_aux_seg=True,
+    bev_seg=True,
+    pv_seg=True,
+    seg_classes=3,
+    feat_down_sample=32,
+    pv_thickness=1)
+model = dict(
+    type='MapTRv2',
+    use_grid_mask=True,
+    video_test_mode=False,
+    pretrained=dict(img='ckpts/resnet50-19c8e357.pth'),
+    img_backbone=dict(
+        type='ResNet',
+        depth=50,
+        num_stages=4,
+        out_indices=(3, ),
+        frozen_stages=1,
+        norm_cfg=dict(type='BN', requires_grad=False),
+        norm_eval=True,
+        style='pytorch'),
+    img_neck=dict(
+        type='FPN',
+        in_channels=[2048],
+        out_channels=256,
+        start_level=0,
+        add_extra_convs='on_output',
+        num_outs=1,
+        relu_before_extra_convs=True),
+    pts_bbox_head=dict(
+        type='MapTRv2Head',
+        bev_h=200,
+        bev_w=100,
+        num_query=900,
+        num_vec_one2one=100,
+        num_vec_one2many=600,
+        k_one2many=6,
+        num_pts_per_vec=20,
+        num_pts_per_gt_vec=20,
+        dir_interval=1,
+        query_embed_type='instance',
+        transform_method='minmax',
+        gt_shift_pts_pattern='v2',
+        num_classes=3,
+        in_channels=256,
+        sync_cls_avg_factor=True,
+        with_box_refine=True,
+        as_two_stage=False,
+        code_size=2,
+        code_weights=[1.0, 1.0, 1.0, 1.0],
+        aux_seg=dict(
+            use_aux_seg=True,
+            bev_seg=True,
+            pv_seg=True,
+            seg_classes=3,
+            feat_down_sample=32,
+            pv_thickness=1),
+        transformer=dict(
+            type='MapTRPerceptionTransformer',
+            rotate_prev_bev=True,
+            use_shift=True,
+            use_can_bus=True,
+            embed_dims=256,
+            encoder=dict(
+                type='BEVFormerEncoder',
+                num_layers=3,
+                pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+                num_points_in_pillar=8,
+                return_intermediate=False,
+                with_height_refine=True,
+                transformerlayers=dict(
+                    type='BEVFormerLayer',
+                    attn_cfgs=[
+                        dict(
+                            type='TemporalSelfAttention',
+                            embed_dims=256,
+                            num_levels=1),
+                        dict(
+                            type='HeightKernelAttention',
+                            pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+                            num_points_in_pillar=8,
+                            attention=dict(
+                                type='MSDeformableAttentionKernel',
+                                embed_dims=256,
+                                num_heads=8,
+                                dilation=1,
+                                kernel_size=(2, 4),
+                                num_levels=1),
+                            embed_dims=256)
+                    ],
+                    feedforward_channels=512,
+                    ffn_dropout=0.1,
+                    operation_order=('self_attn', 'norm', 'cross_attn', 'norm',
+                                     'ffn', 'norm'))),
+            decoder=dict(
+                type='MapTRDecoder',
+                num_layers=6,
+                return_intermediate=True,
+                query_pos_embedding='instance',
+                num_pts_per_vec=20,
+                queru_mask_num=700,
+                transformerlayers=dict(
+                    type='DetrTransformerDecoderLayer',
+                    attn_cfgs=[
+                        dict(
+                            type='MultiheadAttention',
+                            embed_dims=256,
+                            num_heads=8,
+                            dropout=0.1),
+                        dict(
+                            type='InstancePointAttention',
+                            embed_dims=256,
+                            num_levels=1,
+                            num_pts_per_vec=20)
+                    ],
+                    feedforward_channels=512,
+                    ffn_dropout=0.1,
+                    operation_order=('self_attn', 'norm', 'cross_attn', 'norm',
+                                     'ffn', 'norm')))),
+        bbox_coder=dict(
+            type='MapTRNMSFreeCoder',
+            post_center_range=[-20, -35, -20, -35, 20, 35, 20, 35],
+            pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+            max_num=50,
+            voxel_size=[0.15, 0.15, 4.0],
+            num_classes=3),
+        positional_encoding=dict(
+            type='LearnedPositionalEncoding',
+            num_feats=128,
+            row_num_embed=200,
+            col_num_embed=100),
+        loss_cls=dict(
+            type='FocalLoss',
+            use_sigmoid=True,
+            gamma=2.0,
+            alpha=0.25,
+            loss_weight=2.0),
+        loss_bbox=dict(type='L1Loss', loss_weight=0.0),
+        loss_iou=dict(type='GIoULoss', loss_weight=0.0),
+        loss_pts=dict(type='PtsL1Loss', loss_weight=5.0),
+        loss_dir=dict(type='PtsDirCosLoss', loss_weight=0.005),
+        loss_seg=dict(type='SimpleLoss', pos_weight=4.0, loss_weight=1.0),
+        loss_pv_seg=dict(type='SimpleLoss', pos_weight=1.0, loss_weight=2.0)),
+    train_cfg=dict(
+        pts=dict(
+            grid_size=[512, 512, 1],
+            voxel_size=[0.15, 0.15, 4.0],
+            point_cloud_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0],
+            out_size_factor=4,
+            assigner=dict(
+                type='MapTRAssigner',
+                cls_cost=dict(type='FocalLossCost', weight=2.0),
+                reg_cost=dict(
+                    type='BBoxL1Cost', weight=0.0, box_format='xywh'),
+                iou_cost=dict(type='IoUCost', iou_mode='giou', weight=0.0),
+                pts_cost=dict(type='OrderedPtsL1Cost', weight=5),
+                pc_range=[-15.0, -30.0, -2.0, 15.0, 30.0, 2.0]))))
+optimizer = dict(
+    type='AdamW',
+    lr=0.0006,
+    paramwise_cfg=dict(custom_keys=dict(img_backbone=dict(lr_mult=0.1))),
+    weight_decay=0.01)
+optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
+lr_config = dict(
+    policy='CosineAnnealing',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.3333333333333333,
+    min_lr_ratio=0.001)
+total_epochs = 24
+runner = dict(type='EpochBasedRunner', max_epochs=24)
+fp16 = dict(loss_scale=512.0)
+find_unused_parameters = True