- Table 1 compares MM-TTA's fusion model against LiDAR-only baselines.
Table 1's W → K and N → K baselines are taken from CodeMerge's Table 3, and those results are for a LiDAR-only detector. No Adapt and Oracle (line 337-340) are also defined as LiDAR-based models. Neither the table nor its caption says this is for lidar-only or fusion. MM-TTA, by contrast, claims to be TTA for a LiDAR–camera fusion model (BEVFusion). Even if the camera branch is seen only as a way to improve LiDAR features, MM-TTA still consumes images at test time: camera BEV features are fused with LiDAR features before the detection head (Fig. 2), and the view transform is adapted online (Fig. 6). MM-TTA uses more input than the baselines. This leaves two possibilities, and neither supports the paper's claims:
-
Read as a LiDAR-only benchmark, Table 1 says nothing about fusion TTA. If Table 1 is meant to compare TTA methods on LiDAR-only detection, MM-TTA's row doesn't belong in it. The paper then has no comparison against other TTA methods applied to a fusion model, which is the setting it claims to address.
-
Read as a fusion result, the improvement can't be attributed to CMM. The fusion detector has an extra modality and already beats the LiDAR model before any adaptation (Table 8: +1.66 / +2.56 (APBEV/ AP3D) on W → K betwen No Adapt. and BEVFusion). Its final margins are modest: +3.78 / +6.88 AP3D over CodeMerge on W → K / N → K, and only +0.0035 mAP over Tent on W → N. Without baselines run on the same fusion model, the margin can't be split between TTA and the capability of the fusion detector, so the claim that MM-TTA outperforms prior TTA methods is unsupported.
- GCA is limited in novelty, ignores the main source of shift, and has weak supporting evidence.
It's standard range alignment. GCA crops the source point-cloud range and picks which source cameras to keep. Cross-domain 3D detection work does this routinely. UDGA [1] simply converted them to a unified detection range and coordinates. This doesn't justify presenting GCA as a main contribution.
It conflicts with the TTA setting. GCA needs the target's sensor layout and target samples before source pre-training (Section 3.2, Step 1), so a separate source model must be trained for each target domain. Standard TTA, including the MOS / DPO / CodeMerge protocol the paper follows, deploys one source model to unknown target streams.
Its measured benefit is small and inconsistently reported. Before TTA, the GCA-pre-trained BEVFusion beats LiDAR NoAdapt by only +1.66 / +2.56 (APBEV/ AP3D) (Table 8). Yet, Section 4.3 claims GCA gives "nearly 16.23 higher AP3D than other methods at the beginning of the TTA stage" (Fig. 7(c)). No ablation separates GCA from CMM.
-
CMM is close to CodeMerge with the selection rule reversed, and it may reduce to merging recent checkpoints. Most of CMM is inherited. The checkpoint bank and K/L schedule follow MOS. The randomly projected representation follows CodeMerge's fingerprints. The merge rule (Eq. 2) has the same form as CodeMerge's Eq. 11. The actual changes are fingerprints computed from weight changes, selection by a consistency clique instead of by diversity, and similarity-based weights.
-
There is a conceptual conflict (consistency versus diversity) between MM-TTA and previoud methods (MOS, CodeMerge). MOS weights models with the inverse Gram matrix, which deliberately favours diverse, less redundant models. CodeMerge selects by ridge leverage scores, which also favours informative, non−redundant models. MM-TTA's CMM keeps the models that are most consistent with each other and weights them by similarity row sums, which is the opposite principle.
[1] Chang, Gyusam, Jiwon Lee, Donghyun Kim, Jinkyu Kim, Dongwook Lee, Daehyun Ji, Sujin Jang, and Sangpil Kim. "Unified domain generalization and adaptation for multi-view 3d object detection." Advances in Neural Information Processing Systems 37 (2024): 58498-58524.