{"id":"270485b7-57d9-44e3-82b2-f9f5c04df240","arxiv_id":"2507.23778","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Half physics converts kinematic SMPL-X poses into velocities that drive a physics engine, preserving the original motion when contact-free and giving physically correct responses when collisions occur.","lead":"This paper proposes 'half physics', a way to run a standard kinematic 3D body model (SMPL-X) inside a physics simulator by converting each animation frame into per-body-part velocities, so the body follows its original motion unless something blocks it. The approach is learning-free, runs in real time, and aims to make human-scene and human-object interactions physically plausible for VR and embodied AI.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Disabling internal force propagation localizes contact impulses to the struck link, so the claimed physically plausible human-object responses are not established; a momentum-transfer test against full rigid-body physics would settle it.","rationale":"The reader's weakest assumption is that internal dynamics can be ignored without breaking the physical-plausibility claim. I agree and sharpen it: the issue is not just missing muscle actuation but the propagation of contact impulses through the skeleton. In a real impact, a foot's collision with a heavy object changes the motion of the whole leg because joint reaction forces transmit momentum; with those internal forces disabled, the body can only respond locally, and the object trajectory may not match what full-body physics would produce. Since the paper's demonstrations of 'physical' human-object interactions depend on exactly this collision response, the concern is load-bearing. The paper itself flags the limitation in Appendix B.3 and Table 4 shows the tension between tracking fidelity and full dynamics. My proposed concrete test is modular: compare half-physics against a full rigid-body simulation of the same character in the same kick scenario, isolating the impact window where the only difference is internal force propagation. If the test shows agreement, the concern is resolved; if it shows mismatch, the central claim should be revised from 'physically plausible interactions' to 'kinematic fidelity with collision-local corrections.' Because the paper already provides qualitative evidence and the method is learning-free, the conditional verdict remains appropriate; this stress-test does not move the reader's verdict.","tokens_in":17544,"tokens_out":7829,"duration_ms":95407,"concrete_test":"Re-implement the Section 4.3 fast-kick setup in a full rigid-body Featherstone simulation of the same SMPL-X variant with identical masses, joint limits, and inertial properties, but with internal force propagation enabled and joint actuation frozen for a 10-20 ms window around first contact. Run the identical kinematic kick through half-physics and through this full-physics control, then measure (a) ball exit speed and direction and (b) pelvis and thigh recoil after impact. If ball speed or body recoil differs by more than a chosen 20% threshold, half-physics' collision response is not representative of whole-body articulated physics, and the physical-plausibility claim should be scoped to collision-local deformation rather than full physical interaction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central practical claim is that half-physics produces physically plausible interactions while preserving kinematic fidelity. The load-bearing assumption enabling this is the Appendix A decision: 'we disable internal force propagation in the Featherstone pipeline and retain only the system's response to external collisions and frictional contacts.' In real articulated impacts, contact impulses propagate through the skeleton via joint reaction forces: a kick to a heavy ball should decelerate the shank, thigh, and pelvis together, and the effective mass at the foot is much larger than the foot link alone. With internal force propagation disabled, each link's post-collision state is governed mainly by its own inertia and the external impulse, while joint velocities are reset toward kinematic targets on the next frame. Therefore, the 200 kg kicking and suitcase examples (Fig. 5h,i) demonstrate collision-local deformation rather than whole-body physical response, and the object's exit velocity can be wrong even when the visual outcome looks plausible. This is not a novelty dispute but a correctness risk for the 'physical interaction' claim. The paper's own Appendix B.3 concedes that internal actuating dynamics are not modeled, and Table 4 shows that enabling internal forces destroys the near-perfect tracking (MPJPE-g rises to 145.5 mm). Thus the fidelity claim is purchased exactly by suppressing the internal dynamics that make an articulated collision response physically faithful.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces \"half-physics,\" a method to embed the kinematic SMPL-X human model into a physics simulator by converting per-frame positions and rotations into body-part velocities. At each timestep, the body parts are assigned the finite-difference velocity toward the next kinematic target, so that in the absence of contact the simulated pose exactly matches the input, while collisions are handled by the physics engine to block penetration and drive object motion. The authors demonstrate the approach on AIST++ dance tracking, penetration removal on the Trumans HSI dataset, and a range of human-object interactions (grasping, lifting, kicking, dropping) with fully simulated object dynamics. The method is learning-free, runs at roughly 950 fps, and generalizes to arbitrary SMPL-X inputs.","tokens_in":17818,"tokens_out":5696,"duration_ms":59309,"significance":"If the claims are taken at face value, the contribution is practically useful: a simple, training-free, real-time bridge between kinematic human motion and physics simulation, with potential applications in VR/AR, embodied AI, and synthetic data augmentation. The paper honestly lists several limitations in Appendix B.3, including the absence of internal actuation dynamics. However, the central claims of \"physical plausibility\" and \"seamless physical interaction\" are undermined by two methodological choices: the near-zero tracking error in Table 1 is a tautology of the velocity override, and the explicit disabling of internal force propagation in Appendix A removes the articulated-body response that a real collision would produce. These issues are load-bearing for the paper's main contribution and need to be addressed before the claims can be accepted.","major_comments":[{"comment":"The reported MPJPE-g of 0.003 mm is a direct consequence of the control law rather than an empirical tracking result. In Algorithm 1, line 7, the global linear velocity is set to (x_body,t - x_body,t-1)/dt, and joint velocities are set to the angular difference to the target divided by dt. In the absence of collisions, the simulated state must therefore advance exactly to the target pose, up to rounding errors. Comparing this with PHC+, whose agent must contend with inertia and internal dynamics, is an apples-to-oranges comparison. The claim that half-physics reduces tracking error from centimeter scale to negligible is misleading; the correct statement is that the output is identical to the input when no contact occurs. Please reframe this comparison or report fidelity only under contacts.","section":"Section 4.1, Table 1 and Algorithm 1"},{"comment":"Disabling internal force propagation in the Featherstone pipeline eliminates the inertial coupling between links. In a real articulated collision, the contact impulse at the foot is resisted by the effective mass of the whole leg (shank, thigh, pelvis), and joint reaction forces transmit deceleration through the skeleton. With internal forces disabled, the post-impact state of the struck link depends almost entirely on the link's own inertia, and the next-frame velocity override then resets the joint velocities toward the kinematic target. As a result, examples such as Figure 5(h) (200 kg ball) and Figure 5(i) (200 kg suitcase) demonstrate collision-local deformation rather than whole-body physical response, and the object's exit speed may be incorrect even when the visual outcome looks plausible. The paper's own Table 4 shows that enabling internal forces destroys the perfect tracking, indicating that the fidelity claim is purchased exactly by suppressing the dynamics needed for a faithful articulated collision response. I recommend either (i) a quantitative validation of contact impulses and object velocities against a full rigid-body simulator with internal forces enabled, or (ii) a careful restatement of the contribution as 'kinematically driven collision response with physically simulated object dynamics,' with the human body's physical plausibility explicitly qualified.","section":"Appendix A, 'Damping-Free Dynamics and the Neglect of Internal Force'; Table 4"},{"comment":"Passive Joint Stiffness Compensation (PJSC) applies a corrective joint torque, which is itself an internal actuation. The main HOI experiments in Section 4.3 use lambda = 0 (no PJSC), so the claimed 'fingers bend under strain' in Figure 5(i) cannot be attributed to passive stiffness; it must be a result of the velocity override and contact dynamics within the timestep, whose physical meaning is unclear. The paper should clarify what physical mechanism produces the bending, and how PJSC (which re-introduces internal torque) interacts with the 'no internal forces' setting and with the reported penetration and tracking metrics.","section":"Appendix A, Eq. (4) and Section 4.3"},{"comment":"The penetration elimination results are plausible and well presented, but the evaluation only reports zero penetration and does not quantify side effects such as foot sliding, body rotation, or deviation from the intended motion after contacts. Appendix B.3 mentions foot sliding as a possible issue, but the main body of the paper claims the motion remains faithful to the original intent without qualification. A quantitative measure of motion deviation after contact (e.g., global trajectory displacement or joint-angle deviation) would strengthen the HSI evaluation and clarify the trade-off between penetration removal and kinematic fidelity.","section":"Section 4.2 and Figure 4"}],"minor_comments":[{"comment":"The text says 'The results are shown in Table 2' but the comparison results appear in Table 1; Table 2 has the HSI results. Please correct the references.","section":"Section 4.1"},{"comment":"The caption contains the typo 'frameword' instead of 'framework'.","section":"Figure 2 caption"},{"comment":"The notation introduces both global body velocity and per-part velocities, but the superscript convention is used inconsistently; for instance, ˙x_body,0_t is used for the root global velocity, while later in the text the same symbol with index i denotes part velocities. Please clarify the indexing in the equations.","section":"Section 3.2"},{"comment":"PJSC is defined only in Appendix A but is referenced in the main text and Figure 8. Consider defining the acronym at first use or moving a short explanation to Section 3.","section":"Figure 8"},{"comment":"Reference [38] cites the SMPL-H/Embodied Hands model as an arXiv preprint; the published version (ACM Transactions on Graphics, 2022) should be cited if available.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The idea is genuinely simple and appealing, and the paper is clearly written with good qualitative demonstrations. My main concern is that the central claim of 'physically plausible interactions' is not currently supported because of the deliberate disabling of internal forces in the simulator. This is not an incidental detail; it is what makes the perfect tracking possible, and it removes the whole-body articulation that a real human collision would exhibit. The authors could address this by adding a small validation study that compares their contact responses with a full rigid-body simulation (with internal forces) on a simple kick or push scenario, or by rewriting the claims to focus on the object dynamics and contact-driven local deformation rather than whole-body physical plausibility. I would not recommend rejection because the method is useful and the limitation is openly discussed, but the current framing overstates the physical realism of the human side."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is a genuinely useful little system, and the core idea is simple enough to be right. The authors take SMPL-X pose sequences, convert per-frame displacements into per-link velocities, feed them into a Featherstone-based simulator with joint damping and internal force propagation disabled, and get collision responses (penetration blocking, object motion) while preserving the kinematic input almost exactly. That is a clean, training-free bridge for data augmentation, VR avatars, and embodied AI experiments. The Passive Joint Stiffness Compensation trick is a nice touch. I believe the method works as described.\n\nThe soft spots are real but not fatal. First, the Table 1 comparison to PHC+ is largely meaningless. HP's MPJPE of 3 micrometers is a tautology: the velocity law is defined as the finite difference to the target, so in the absence of collisions the output is the input by construction. It is not a learned tracker and should not be marketed as one. The paper would be better off framing this as 'fidelity preservation' and dropping the PHC+ comparison or restricting it to success rate.\n\nSecond, the 'physically plausible interaction' claim overreaches. Appendix A says they disable internal force propagation in the Featherstone pipeline and keep only external collision response. In real articulated impacts, contact impulses propagate through joints; a kick to a 200kg ball should decelerate the shank, thigh, and pelvis together. With internal forces off, the effective mass at the foot is just the foot link, so the post-collision body pose is collision-local deformation rather than whole-body physics. Table 4 confirms the trade-off: turning internal forces back on destroys the tracking fidelity (145mm error). So the fidelity is purchased exactly by suppressing the dynamics that make articulated collisions physically faithful. The authors concede in B.3 that they do not model internal actuation, but the main text's 'physical plausibility' language is stronger than what the pipeline delivers.\n\nNone of this sinks the paper. The HSI penetration results are straightforward and useful, the HOI demos are suggestive, and the method is real-time and generalizes across body shapes and motions. What it needs is a revision that (1) reframes the tracking comparison, (2) tempers 'physical' to 'collision-aware' or adds a momentum-transfer analysis against full rigid-body physics, and (3) includes a quantitative HOI baseline if possible. As is, it deserves a serious referee. I'd bring it to reading group and would cite it for its practical bridge value.","headline":"Useful, simple kinematic-to-physics bridge for SMPL-X, but the PHC+ comparison is circular and the 'physical plausibility' claim outruns the disabled-internal-forces pipeline.","tokens_in":18318,"tokens_out":2455,"would_cite":true,"duration_ms":25245,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A kinematic 3D human motion can be made physically interactive with no learning by replacing per-frame poses with per-body-part velocities, so that the body tracks its input exactly until a collision, at which point physics takes over.","keywords":["half physics","kinematic-to-physics transfer","velocity override","SMPL-X","human-scene interaction","human-object interaction","physics simulation","penetration resolution"],"falsifier":"Run a motion-capture experiment in which a human kicks a 0.5 kg ball and a 200 kg ball with the same intended motion, and compare post-contact joint-angle trajectories with half-physics predictions; if the human's joints show muscle-driven braking or pre-impact stiffening that the simulator, which disables internal forces, cannot reproduce, the claim that physical plausibility is preserved after collisions is falsified.","tokens_in":1480,"feed_emoji":"🏃","tokens_out":1729,"duration_ms":76145,"temperature":0.7,"pith_summary":"The paper tries to establish that a purely kinematic 3D human motion sequence — positions and joint rotations per frame, with no forces or velocities — can be run inside a physics simulation by converting each frame's target into a velocity. It claims this \"half physics\" scheme preserves the original motion to machine precision when nothing is touched, lets the simulator resolve collisions and penetration when something is, and needs no training, no reward engineering, and no per-motion or per-body retraining. If true, any kinematic human model used in VR, animation, or embodied AI can gain physically consistent reactions — stopping at walls, lifting, dropping, and kicking objects — purely as an emergent consequence of running the same poses through a physics engine at real-time speed.","feed_headline":"Velocity trick gives kinematic human motion real collisions","feed_subtitle":"Per-frame poses become velocities, so the body tracks its input and only physics changes the outcome.","key_machinery":"The central mechanism is \"half physics\": a velocity-override bridge between kinematics and dynamics. Instead of applying joint torques, the method computes the linear velocity of each body part as the finite difference between the current position and the next kinematic target, applies the same finite-difference and spherical-interpolation scheme to global and joint angular velocities, and lets a simulator step the system forward. The body thus moves ballistically toward the kinematic target when nothing touches it, and the engine's collision and friction solver takes over when something does; internal-force and damping terms are switched off because the kinematic input is taken to already carry the muscular coordination, and an optional passive stiffness compensation torque can pull the pose back toward the intended trajectory after contact.","core_discovery":"The paper's central claim is that a kinematic SMPL-X motion can be embedded in a physics engine with no learning by directly overriding per-body-part velocities instead of driving joints with torques. On dance motions, this achieves tracking error of about 3 micrometers versus centimeters for a trained physical tracker, and 100% success without falls; on scene interactions it reduces penetration from 7.9% of vertices to zero; and on object interactions it produces emergent responses — grasp, lift, drop, kick — that depend correctly on object mass and friction, without any predicted object trajectories. This is presented as a faithful transfer: absent collisions the output equals the kinematic input, and upon collisions the physics engine dictates the interaction.","pith_inferences":["If the central claim holds, half physics offers a cheap post-processing layer for generative motion models: sample kinematic poses, then let the simulator decide what happens at contact, with object state as a free byproduct of the interaction.","The method's own limitation text notes that internal muscle-skeletal dynamics are not modeled, so its \"physical plausibility\" is about external-contact behavior, not biomechanical fidelity; a biomechanical validation study would be the decisive test of that scope.","The stiffness parameter lambda suggests a controllable trade-off between preserving local joint poses and allowing whole-body displacement; a natural extension would be to fit lambda per motion or per body style from human muscle-tension data.","Because the approach does not semantically correct the motion, it would not by itself fix an input that walks into a newly inserted wall; combining it with a motion planner or semantic scene-aware generator is a promising division of labor."],"forward_implications":["Any kinematic SMPL-X motion sequence can be replayed inside a physics engine with zero training, preserving the original poses while blocking penetration against static scenes.","Object responses in human-object interaction emerge from mass, friction, and contact rather than from learned or scripted object trajectories, so physical factors can be varied to generate new interaction outcomes.","Tracking fidelity shifts from centimeter-scale errors (typical of torque-driven learned trackers) to near machine precision, making the method suitable for applications such as VR where exact alignment with user motion matters.","The same half-physics layer can augment interaction datasets with physically realistic labels, such as what happens when a ball is very heavy or a surface is very slippery.","Because the conversion is a simple velocity computation, the approach can be ported to other physics simulators that expose per-link velocity interfaces."],"supporting_citations":[{"why":"Defines the parametric body and pose representation whose kinematics the method converts to velocities.","marker":"[33]"},{"why":"Provides the articulated-body dynamics used by the simulator to step states and resolve collisions and friction.","marker":"[5]"},{"why":"The strongest learned physical-tracker baseline, used for the fidelity and success-rate comparison.","marker":"[24]"},{"why":"The dance-motion test set on which tracking error and fall rate are measured.","marker":"[18]"},{"why":"The human-scene interaction benchmark from which SMPL-X motions with penetrations are taken.","marker":"[13]"},{"why":"Kinematic whole-body grasping sequences used as input for object-interaction demonstrations.","marker":"[45]"},{"why":"Kinematic human-object interaction sequences used to show emergent, contact-consistent object motion.","marker":"[17]"},{"why":"Motion-capture archive from which the kicking motions used in object-interaction experiments are drawn.","marker":"[29]"}],"fun_headline_variants":["Learning-free physics gives kinematic human motion real collisions","No-training trick adds physics to kinematic human motion","Velocity override yields physical interactions for kinematic SMPL-X","Half-physics: real-time kinematic motion that collides correctly","Kinematic humans get physical response with zero training"],"cache_read_input_tokens":20480,"weakest_assumption_plain":"The load-bearing assumption is that a kinematic motion sequence already encodes the outcome of the body's muscular coordination, so the simulator can safely ignore internal joint forces and damping and respond only to external collisions; if real muscle dynamics matter right after impact, the post-collision body responses are not biomechanically faithful.","fun_headline_variants_meta":{"raw":{"variants":["Learning-free physics gives kinematic human motion real collisions","No-training trick adds physics to kinematic human motion","Velocity override yields physical interactions for kinematic SMPL-X","Half-physics: real-time kinematic motion that collides correctly","Kinematic humans get physical response with zero training"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000241,"raw_usage":{"total_tokens":1476,"prompt_tokens":851,"completion_tokens":625,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":467,"completion_tokens_details":{"reasoning_tokens":550}},"tokens_in":467,"tokens_out":625,"duration_ms":6798,"temperature":1.0,"reasoning_tokens":550,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T10:25:01.266835+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a motion-capture experiment in which a human kicks a 0.5 kg ball and a 200 kg ball with the same intended motion, and compare post-contact joint-angle trajectories with half-physics predictions; if the human's joints show muscle-driven braking or pre-impact stiffening that the simulator, which disables internal forces, cannot reproduce, the claim that physical plausibility is preserved after collisions is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the parametric body and pose representation whose kinematics the method converts to velocities."},{"cited_title":"The calculation of robot dynamics using articulated-body inertias","cited_arxiv_id":null,"evidence_quote":"Provides the articulated-body dynamics used by the simulator to step states and resolve collisions and friction."},{"cited_title":"Universal humanoid motion representations for physics-based control","cited_arxiv_id":null,"evidence_quote":"The strongest learned physical-tracker baseline, used for the fidelity and success-rate comparison."},{"cited_title":"Ai choreographer: Music conditioned 3d dance generation with aist++","cited_arxiv_id":null,"evidence_quote":"The dance-motion test set on which tracking error and fall rate are measured."},{"cited_title":"Scaling up dynamic human-scene interaction modeling","cited_arxiv_id":null,"evidence_quote":"The human-scene interaction benchmark from which SMPL-X motions with penetrations are taken."},{"cited_title":"Black, and Dimitrios Tzionas","cited_arxiv_id":null,"evidence_quote":"Kinematic whole-body grasping sequences used as input for object-interaction demonstrations."},{"cited_title":"Karen Liu","cited_arxiv_id":null,"evidence_quote":"Kinematic human-object interaction sequences used to show emergent, contact-consistent object motion."},{"cited_title":"Amass: Archive of motion capture as surface shapes","cited_arxiv_id":null,"evidence_quote":"Motion-capture archive from which the kicking motions used in object-interaction experiments are drawn."}],"review_version":1}